├── README.md ├── efficient_architecture_llm.md ├── efficient_moe.md ├── efficient_plm └── readme.md ├── efficient_training.md ├── figures ├── 2ssp.png ├── AFPQ.png ├── ALPS.png ├── AMXFP4.png ├── ANPD.png ├── APT.png ├── AQAS.png ├── ATKD.png ├── AWE.png ├── AnyPrecisionLLM.png ├── ApiQ.png ├── AutoCompressor.png ├── AutoMixQ.png ├── BOND.png ├── BPD.png ├── BabyLLaMA.png ├── BiTA.png ├── CCEMF.png ├── CCM.png ├── CCRAG_survey.png ├── CD.png ├── CIM.png ├── CLA.png ├── CPET.png ├── CSDrafting.png ├── CacheGen.png ├── ChunkAttention.png ├── CoMD.png ├── CoScript.png ├── CoT-Max.png ├── ConsistentEE.png ├── CrossTokenizer.png ├── DB-LLM.png ├── DCP.png ├── DGQ.png ├── DISCO_2.png ├── DMC.png ├── DSI.png ├── DSOT.png ├── DaSS.png ├── DajeVu.png ├── DecoQuant.png ├── DeltaNet.png ├── DistillSpec.png ├── Distill_step_by_step.png ├── Doctor.png ├── EBFT.png ├── ESPACE.png ├── EXAQ.png ├── EdgeQAT.png ├── EntropyRank.png ├── EoTD.png ├── FDT.png ├── FP6-LLM.png ├── FPTQ.png ├── FREE.png ├── FalsePromise.png ├── FastCoT.png ├── FastGen.png ├── Flab-Pruner.png ├── FlashAttention2.png ├── FlashDecoding++.png ├── FlashLLM.png ├── GBLM-Pruner.png ├── GCD.png ├── GKD.png ├── GPT-Zip.png ├── GPT4AIGChip.png ├── GPTQ.png ├── GuidedQuant.png ├── HGRN.png ├── HO-Ring.png ├── I-LLM.png ├── ICAE.png ├── IDP.png ├── KVQuant.png ├── KV_compression.png ├── L4Q.png ├── LCKV.png ├── LESS.png ├── LITE.png ├── LLM-FP4.png ├── LLM-KICK.png ├── LLM-MPQ.png ├── LLM-QAT.png ├── LLM-ROM.png ├── LLM-shearing.png ├── LLMZip.png ├── LLM_MoT_cascade.png ├── LPLR.png ├── LQ-LoRA.png ├── LRC.png ├── LTE.png ├── LaCo.png ├── LayerSkip.png ├── LinguGKD.png ├── LoMA.png ├── LoNAS.png ├── LoRAPrune.png ├── LoRAShear.png ├── LoRC.png ├── LoRD.png ├── LoSparse.png ├── LoftQ.png ├── Lookahead.png ├── MBPP.png ├── MCQ.png ├── MLFS.png ├── MOE-Infinity.png ├── Mamba-Shedder.png ├── Mamba.png ├── Mamba2.png ├── MiniMA.png ├── MixDistill.png ├── MoFQ.png ├── MobileLLM.png ├── ModelSpecialization.png ├── ModuLoRA.png ├── MultiPruner.png ├── NASH.png ├── NormTweaking.png ├── OCaTS.png ├── OWQ.png ├── OdysseyLLM.png ├── OmniKV.png ├── OneBit.png ├── OpenBA.png ├── P-RGE.png ├── PAT.png ├── PB-LLM.png ├── PEQA.png ├── PERP.png ├── PaD.png ├── PalmBench.png ├── ProPD.png ├── PromptKD.png ├── PromptMix.png ├── PruningAccuracyPredictor.png ├── PyramidInfer.png ├── Q-BaRA.png ├── QFT.png ├── QJL.png ├── QLLM.png ├── QT.png ├── QuIP.png ├── QuIP_sign.png ├── QuantEase.png ├── QuantizationStrategies.png ├── QuantizedEmpirical.png ├── RDRec.png ├── RIA.png ├── RLCD.png ├── RSD.png ├── ResQ.png ├── RetNet.png ├── RetriKT.png ├── SCOTT.png ├── SCoTD.png ├── SDFT.png ├── SFSD-LLM.png ├── SKVQ.png ├── SLEB.png ├── SMoA.png ├── SPEED.png ├── SQFT.png ├── ScalingEfficientLLM.png ├── Sci-COT.png ├── Scissorhands.png ├── Setwise.png ├── Shears.png ├── ShortenedLLaMA.png ├── SiDA.png ├── SignRound.png ├── SkipDecode.png ├── SliceGPT.png ├── SmoothQuant+.png ├── SoT.png ├── SpQR.png ├── SparQ.png ├── SparseFrontier.png ├── SpeculativeStreaming.png ├── SquareHead.png ├── SqueezeLLM.png ├── StagedSpec.png ├── TAPIR.png ├── TCRA-LLM.png ├── TEQ.png ├── TT-SVD.png ├── TTT.png ├── Tandem.png ├── Teach_Small_LM_COT.png ├── TeacherFreeLLM.png ├── Titans.png ├── UniversalNER.png ├── VPTQ.png ├── VTrans.png ├── WKVQuant.png ├── YOCO.png ├── ZeroQuant-FP.png ├── abq-llm.png ├── adakv.png ├── admm.png ├── agile.png ├── atom.png ├── bolaco.png ├── bonsai.png ├── cacheblend-pic.png ├── cachecraft.png ├── cachefocus.png ├── cachegen-snapshot.png ├── cacheprior.png ├── canonical_tensor_decomposition.png ├── chai.png ├── clover.png ├── compress_rank.png ├── compresso.png ├── concept_RAG.png ├── copy_icl.png ├── cross-layer-kv.png ├── disco.png ├── distill_event.png ├── dynmoe.png ├── e-sparse.png ├── epic_img.png ├── eval_safety.png ├── evopress.png ├── exflow.png ├── exploiting_llm_quantization.png ├── fast-2-bit.png ├── flashattention3.png ├── high_sparsity_pretraining.png ├── homer.png ├── hybridLLM.png ├── hymba.png ├── impossible_distillation.png ├── junk_DNA.png ├── kcache.png ├── kd_close_source.png ├── kvzip.png ├── learn-to-reason.png ├── lillama.png ├── llm_pruner.png ├── llma.png ├── longctx_bench.png ├── longllmlingua.png ├── mamba_drafters.png ├── megalodon.png ├── minicache.png ├── minillm.png ├── mixtral_offloading.png ├── multiflow.png ├── neuroal.png ├── nugget2D.png ├── omniquant.png ├── outliersuppression.png ├── parrot.png ├── preble.png ├── pv-tuning.png ├── qllm-eval.png ├── qlora.png ├── quantized-lm-confidence.png ├── recall_and_icl.png ├── reform.png ├── relu2wins.png ├── relufication.png ├── rethink-AKL.png ├── selective-cloud-llm-assistance.png ├── selective_context.png ├── sensitivity_sparse.png ├── simba.png ├── sparsegpt.png ├── speccascade.png ├── spinquant.png ├── stand.png ├── survey_xunyu.png ├── switch.png ├── switchhead.png ├── unrolled.png ├── watermark_quant.png ├── zephyr.png ├── zeroquant-6bit.png ├── zeroquant-v2.png └── zipcache.png ├── generate_item.py ├── hardware.md ├── inference_acceleration.md ├── knowledge_distillation.md ├── kv_cache_compression.md ├── leaderboard.md ├── low_rank_decomposition.md ├── project └── readme.md ├── pruning.md ├── quantization.md ├── survey.md ├── text_compression.md └── tuning.md /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/README.md -------------------------------------------------------------------------------- /efficient_architecture_llm.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/efficient_architecture_llm.md -------------------------------------------------------------------------------- /efficient_moe.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/efficient_moe.md -------------------------------------------------------------------------------- /efficient_plm/readme.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/efficient_plm/readme.md -------------------------------------------------------------------------------- /efficient_training.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/efficient_training.md -------------------------------------------------------------------------------- /figures/2ssp.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/2ssp.png -------------------------------------------------------------------------------- /figures/AFPQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/AFPQ.png -------------------------------------------------------------------------------- /figures/ALPS.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ALPS.png -------------------------------------------------------------------------------- /figures/AMXFP4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/AMXFP4.png -------------------------------------------------------------------------------- /figures/ANPD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ANPD.png -------------------------------------------------------------------------------- /figures/APT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/APT.png -------------------------------------------------------------------------------- /figures/AQAS.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/AQAS.png -------------------------------------------------------------------------------- /figures/ATKD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ATKD.png -------------------------------------------------------------------------------- /figures/AWE.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/AWE.png -------------------------------------------------------------------------------- /figures/AnyPrecisionLLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/AnyPrecisionLLM.png -------------------------------------------------------------------------------- /figures/ApiQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ApiQ.png -------------------------------------------------------------------------------- /figures/AutoCompressor.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/AutoCompressor.png -------------------------------------------------------------------------------- /figures/AutoMixQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/AutoMixQ.png -------------------------------------------------------------------------------- /figures/BOND.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/BOND.png -------------------------------------------------------------------------------- /figures/BPD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/BPD.png -------------------------------------------------------------------------------- /figures/BabyLLaMA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/BabyLLaMA.png -------------------------------------------------------------------------------- /figures/BiTA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/BiTA.png -------------------------------------------------------------------------------- /figures/CCEMF.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CCEMF.png -------------------------------------------------------------------------------- /figures/CCM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CCM.png -------------------------------------------------------------------------------- /figures/CCRAG_survey.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CCRAG_survey.png -------------------------------------------------------------------------------- /figures/CD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CD.png -------------------------------------------------------------------------------- /figures/CIM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CIM.png -------------------------------------------------------------------------------- /figures/CLA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CLA.png -------------------------------------------------------------------------------- /figures/CPET.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CPET.png -------------------------------------------------------------------------------- /figures/CSDrafting.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CSDrafting.png -------------------------------------------------------------------------------- /figures/CacheGen.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CacheGen.png -------------------------------------------------------------------------------- /figures/ChunkAttention.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ChunkAttention.png -------------------------------------------------------------------------------- /figures/CoMD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CoMD.png -------------------------------------------------------------------------------- /figures/CoScript.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CoScript.png -------------------------------------------------------------------------------- /figures/CoT-Max.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CoT-Max.png -------------------------------------------------------------------------------- /figures/ConsistentEE.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ConsistentEE.png -------------------------------------------------------------------------------- /figures/CrossTokenizer.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/CrossTokenizer.png -------------------------------------------------------------------------------- /figures/DB-LLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DB-LLM.png -------------------------------------------------------------------------------- /figures/DCP.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DCP.png -------------------------------------------------------------------------------- /figures/DGQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DGQ.png -------------------------------------------------------------------------------- /figures/DISCO_2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DISCO_2.png -------------------------------------------------------------------------------- /figures/DMC.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DMC.png -------------------------------------------------------------------------------- /figures/DSI.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DSI.png -------------------------------------------------------------------------------- /figures/DSOT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DSOT.png -------------------------------------------------------------------------------- /figures/DaSS.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DaSS.png -------------------------------------------------------------------------------- /figures/DajeVu.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DajeVu.png -------------------------------------------------------------------------------- /figures/DecoQuant.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DecoQuant.png -------------------------------------------------------------------------------- /figures/DeltaNet.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DeltaNet.png -------------------------------------------------------------------------------- /figures/DistillSpec.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/DistillSpec.png -------------------------------------------------------------------------------- /figures/Distill_step_by_step.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Distill_step_by_step.png -------------------------------------------------------------------------------- /figures/Doctor.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Doctor.png -------------------------------------------------------------------------------- /figures/EBFT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/EBFT.png -------------------------------------------------------------------------------- /figures/ESPACE.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ESPACE.png -------------------------------------------------------------------------------- /figures/EXAQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/EXAQ.png -------------------------------------------------------------------------------- /figures/EdgeQAT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/EdgeQAT.png -------------------------------------------------------------------------------- /figures/EntropyRank.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/EntropyRank.png -------------------------------------------------------------------------------- /figures/EoTD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/EoTD.png -------------------------------------------------------------------------------- /figures/FDT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/FDT.png -------------------------------------------------------------------------------- /figures/FP6-LLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/FP6-LLM.png -------------------------------------------------------------------------------- /figures/FPTQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/FPTQ.png -------------------------------------------------------------------------------- /figures/FREE.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/FREE.png -------------------------------------------------------------------------------- /figures/FalsePromise.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/FalsePromise.png -------------------------------------------------------------------------------- /figures/FastCoT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/FastCoT.png -------------------------------------------------------------------------------- /figures/FastGen.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/FastGen.png -------------------------------------------------------------------------------- /figures/Flab-Pruner.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Flab-Pruner.png -------------------------------------------------------------------------------- /figures/FlashAttention2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/FlashAttention2.png -------------------------------------------------------------------------------- /figures/FlashDecoding++.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/FlashDecoding++.png -------------------------------------------------------------------------------- /figures/FlashLLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/FlashLLM.png -------------------------------------------------------------------------------- /figures/GBLM-Pruner.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/GBLM-Pruner.png -------------------------------------------------------------------------------- /figures/GCD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/GCD.png -------------------------------------------------------------------------------- /figures/GKD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/GKD.png -------------------------------------------------------------------------------- /figures/GPT-Zip.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/GPT-Zip.png -------------------------------------------------------------------------------- /figures/GPT4AIGChip.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/GPT4AIGChip.png -------------------------------------------------------------------------------- /figures/GPTQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/GPTQ.png -------------------------------------------------------------------------------- /figures/GuidedQuant.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/GuidedQuant.png -------------------------------------------------------------------------------- /figures/HGRN.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/HGRN.png -------------------------------------------------------------------------------- /figures/HO-Ring.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/HO-Ring.png -------------------------------------------------------------------------------- /figures/I-LLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/I-LLM.png -------------------------------------------------------------------------------- /figures/ICAE.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ICAE.png -------------------------------------------------------------------------------- /figures/IDP.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/IDP.png -------------------------------------------------------------------------------- /figures/KVQuant.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/KVQuant.png -------------------------------------------------------------------------------- /figures/KV_compression.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/KV_compression.png -------------------------------------------------------------------------------- /figures/L4Q.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/L4Q.png -------------------------------------------------------------------------------- /figures/LCKV.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LCKV.png -------------------------------------------------------------------------------- /figures/LESS.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LESS.png -------------------------------------------------------------------------------- /figures/LITE.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LITE.png -------------------------------------------------------------------------------- /figures/LLM-FP4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LLM-FP4.png -------------------------------------------------------------------------------- /figures/LLM-KICK.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LLM-KICK.png -------------------------------------------------------------------------------- /figures/LLM-MPQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LLM-MPQ.png -------------------------------------------------------------------------------- /figures/LLM-QAT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LLM-QAT.png -------------------------------------------------------------------------------- /figures/LLM-ROM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LLM-ROM.png -------------------------------------------------------------------------------- /figures/LLM-shearing.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LLM-shearing.png -------------------------------------------------------------------------------- /figures/LLMZip.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LLMZip.png -------------------------------------------------------------------------------- /figures/LLM_MoT_cascade.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LLM_MoT_cascade.png -------------------------------------------------------------------------------- /figures/LPLR.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LPLR.png -------------------------------------------------------------------------------- /figures/LQ-LoRA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LQ-LoRA.png -------------------------------------------------------------------------------- /figures/LRC.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LRC.png -------------------------------------------------------------------------------- /figures/LTE.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LTE.png -------------------------------------------------------------------------------- /figures/LaCo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LaCo.png -------------------------------------------------------------------------------- /figures/LayerSkip.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LayerSkip.png -------------------------------------------------------------------------------- /figures/LinguGKD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LinguGKD.png -------------------------------------------------------------------------------- /figures/LoMA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LoMA.png -------------------------------------------------------------------------------- /figures/LoNAS.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LoNAS.png -------------------------------------------------------------------------------- /figures/LoRAPrune.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LoRAPrune.png -------------------------------------------------------------------------------- /figures/LoRAShear.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LoRAShear.png -------------------------------------------------------------------------------- /figures/LoRC.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LoRC.png -------------------------------------------------------------------------------- /figures/LoRD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LoRD.png -------------------------------------------------------------------------------- /figures/LoSparse.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LoSparse.png -------------------------------------------------------------------------------- /figures/LoftQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/LoftQ.png -------------------------------------------------------------------------------- /figures/Lookahead.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Lookahead.png -------------------------------------------------------------------------------- /figures/MBPP.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/MBPP.png -------------------------------------------------------------------------------- /figures/MCQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/MCQ.png -------------------------------------------------------------------------------- /figures/MLFS.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/MLFS.png -------------------------------------------------------------------------------- /figures/MOE-Infinity.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/MOE-Infinity.png -------------------------------------------------------------------------------- /figures/Mamba-Shedder.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Mamba-Shedder.png -------------------------------------------------------------------------------- /figures/Mamba.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Mamba.png -------------------------------------------------------------------------------- /figures/Mamba2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Mamba2.png -------------------------------------------------------------------------------- /figures/MiniMA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/MiniMA.png -------------------------------------------------------------------------------- /figures/MixDistill.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/MixDistill.png -------------------------------------------------------------------------------- /figures/MoFQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/MoFQ.png -------------------------------------------------------------------------------- /figures/MobileLLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/MobileLLM.png -------------------------------------------------------------------------------- /figures/ModelSpecialization.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ModelSpecialization.png -------------------------------------------------------------------------------- /figures/ModuLoRA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ModuLoRA.png -------------------------------------------------------------------------------- /figures/MultiPruner.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/MultiPruner.png -------------------------------------------------------------------------------- /figures/NASH.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/NASH.png -------------------------------------------------------------------------------- /figures/NormTweaking.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/NormTweaking.png -------------------------------------------------------------------------------- /figures/OCaTS.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/OCaTS.png -------------------------------------------------------------------------------- /figures/OWQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/OWQ.png -------------------------------------------------------------------------------- /figures/OdysseyLLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/OdysseyLLM.png -------------------------------------------------------------------------------- /figures/OmniKV.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/OmniKV.png -------------------------------------------------------------------------------- /figures/OneBit.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/OneBit.png -------------------------------------------------------------------------------- /figures/OpenBA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/OpenBA.png -------------------------------------------------------------------------------- /figures/P-RGE.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/P-RGE.png -------------------------------------------------------------------------------- /figures/PAT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/PAT.png -------------------------------------------------------------------------------- /figures/PB-LLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/PB-LLM.png -------------------------------------------------------------------------------- /figures/PEQA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/PEQA.png -------------------------------------------------------------------------------- /figures/PERP.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/PERP.png -------------------------------------------------------------------------------- /figures/PaD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/PaD.png -------------------------------------------------------------------------------- /figures/PalmBench.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/PalmBench.png -------------------------------------------------------------------------------- /figures/ProPD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ProPD.png -------------------------------------------------------------------------------- /figures/PromptKD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/PromptKD.png -------------------------------------------------------------------------------- /figures/PromptMix.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/PromptMix.png -------------------------------------------------------------------------------- /figures/PruningAccuracyPredictor.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/PruningAccuracyPredictor.png -------------------------------------------------------------------------------- /figures/PyramidInfer.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/PyramidInfer.png -------------------------------------------------------------------------------- /figures/Q-BaRA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Q-BaRA.png -------------------------------------------------------------------------------- /figures/QFT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/QFT.png -------------------------------------------------------------------------------- /figures/QJL.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/QJL.png -------------------------------------------------------------------------------- /figures/QLLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/QLLM.png -------------------------------------------------------------------------------- /figures/QT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/QT.png -------------------------------------------------------------------------------- /figures/QuIP.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/QuIP.png -------------------------------------------------------------------------------- /figures/QuIP_sign.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/QuIP_sign.png -------------------------------------------------------------------------------- /figures/QuantEase.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/QuantEase.png -------------------------------------------------------------------------------- /figures/QuantizationStrategies.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/QuantizationStrategies.png -------------------------------------------------------------------------------- /figures/QuantizedEmpirical.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/QuantizedEmpirical.png -------------------------------------------------------------------------------- /figures/RDRec.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/RDRec.png -------------------------------------------------------------------------------- /figures/RIA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/RIA.png -------------------------------------------------------------------------------- /figures/RLCD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/RLCD.png -------------------------------------------------------------------------------- /figures/RSD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/RSD.png -------------------------------------------------------------------------------- /figures/ResQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ResQ.png -------------------------------------------------------------------------------- /figures/RetNet.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/RetNet.png -------------------------------------------------------------------------------- /figures/RetriKT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/RetriKT.png -------------------------------------------------------------------------------- /figures/SCOTT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SCOTT.png -------------------------------------------------------------------------------- /figures/SCoTD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SCoTD.png -------------------------------------------------------------------------------- /figures/SDFT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SDFT.png -------------------------------------------------------------------------------- /figures/SFSD-LLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SFSD-LLM.png -------------------------------------------------------------------------------- /figures/SKVQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SKVQ.png -------------------------------------------------------------------------------- /figures/SLEB.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SLEB.png -------------------------------------------------------------------------------- /figures/SMoA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SMoA.png -------------------------------------------------------------------------------- /figures/SPEED.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SPEED.png -------------------------------------------------------------------------------- /figures/SQFT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SQFT.png -------------------------------------------------------------------------------- /figures/ScalingEfficientLLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ScalingEfficientLLM.png -------------------------------------------------------------------------------- /figures/Sci-COT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Sci-COT.png -------------------------------------------------------------------------------- /figures/Scissorhands.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Scissorhands.png -------------------------------------------------------------------------------- /figures/Setwise.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Setwise.png -------------------------------------------------------------------------------- /figures/Shears.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Shears.png -------------------------------------------------------------------------------- /figures/ShortenedLLaMA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ShortenedLLaMA.png -------------------------------------------------------------------------------- /figures/SiDA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SiDA.png -------------------------------------------------------------------------------- /figures/SignRound.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SignRound.png -------------------------------------------------------------------------------- /figures/SkipDecode.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SkipDecode.png -------------------------------------------------------------------------------- /figures/SliceGPT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SliceGPT.png -------------------------------------------------------------------------------- /figures/SmoothQuant+.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SmoothQuant+.png -------------------------------------------------------------------------------- /figures/SoT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SoT.png -------------------------------------------------------------------------------- /figures/SpQR.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SpQR.png -------------------------------------------------------------------------------- /figures/SparQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SparQ.png -------------------------------------------------------------------------------- /figures/SparseFrontier.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SparseFrontier.png -------------------------------------------------------------------------------- /figures/SpeculativeStreaming.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SpeculativeStreaming.png -------------------------------------------------------------------------------- /figures/SquareHead.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SquareHead.png -------------------------------------------------------------------------------- /figures/SqueezeLLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/SqueezeLLM.png -------------------------------------------------------------------------------- /figures/StagedSpec.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/StagedSpec.png -------------------------------------------------------------------------------- /figures/TAPIR.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/TAPIR.png -------------------------------------------------------------------------------- /figures/TCRA-LLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/TCRA-LLM.png -------------------------------------------------------------------------------- /figures/TEQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/TEQ.png -------------------------------------------------------------------------------- /figures/TT-SVD.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/TT-SVD.png -------------------------------------------------------------------------------- /figures/TTT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/TTT.png -------------------------------------------------------------------------------- /figures/Tandem.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Tandem.png -------------------------------------------------------------------------------- /figures/Teach_Small_LM_COT.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Teach_Small_LM_COT.png -------------------------------------------------------------------------------- /figures/TeacherFreeLLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/TeacherFreeLLM.png -------------------------------------------------------------------------------- /figures/Titans.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/Titans.png -------------------------------------------------------------------------------- /figures/UniversalNER.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/UniversalNER.png -------------------------------------------------------------------------------- /figures/VPTQ.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/VPTQ.png -------------------------------------------------------------------------------- /figures/VTrans.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/VTrans.png -------------------------------------------------------------------------------- /figures/WKVQuant.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/WKVQuant.png -------------------------------------------------------------------------------- /figures/YOCO.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/YOCO.png -------------------------------------------------------------------------------- /figures/ZeroQuant-FP.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/ZeroQuant-FP.png -------------------------------------------------------------------------------- /figures/abq-llm.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/abq-llm.png -------------------------------------------------------------------------------- /figures/adakv.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/adakv.png -------------------------------------------------------------------------------- /figures/admm.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/admm.png -------------------------------------------------------------------------------- /figures/agile.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/agile.png -------------------------------------------------------------------------------- /figures/atom.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/atom.png -------------------------------------------------------------------------------- /figures/bolaco.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/bolaco.png -------------------------------------------------------------------------------- /figures/bonsai.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/bonsai.png -------------------------------------------------------------------------------- /figures/cacheblend-pic.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/cacheblend-pic.png -------------------------------------------------------------------------------- /figures/cachecraft.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/cachecraft.png -------------------------------------------------------------------------------- /figures/cachefocus.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/cachefocus.png -------------------------------------------------------------------------------- /figures/cachegen-snapshot.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/cachegen-snapshot.png -------------------------------------------------------------------------------- /figures/cacheprior.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/cacheprior.png -------------------------------------------------------------------------------- /figures/canonical_tensor_decomposition.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/canonical_tensor_decomposition.png -------------------------------------------------------------------------------- /figures/chai.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/chai.png -------------------------------------------------------------------------------- /figures/clover.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/clover.png -------------------------------------------------------------------------------- /figures/compress_rank.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/compress_rank.png -------------------------------------------------------------------------------- /figures/compresso.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/compresso.png -------------------------------------------------------------------------------- /figures/concept_RAG.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/concept_RAG.png -------------------------------------------------------------------------------- /figures/copy_icl.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/copy_icl.png -------------------------------------------------------------------------------- /figures/cross-layer-kv.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/cross-layer-kv.png -------------------------------------------------------------------------------- /figures/disco.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/disco.png -------------------------------------------------------------------------------- /figures/distill_event.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/distill_event.png -------------------------------------------------------------------------------- /figures/dynmoe.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/dynmoe.png -------------------------------------------------------------------------------- /figures/e-sparse.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/e-sparse.png -------------------------------------------------------------------------------- /figures/epic_img.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/epic_img.png -------------------------------------------------------------------------------- /figures/eval_safety.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/eval_safety.png -------------------------------------------------------------------------------- /figures/evopress.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/evopress.png -------------------------------------------------------------------------------- /figures/exflow.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/exflow.png -------------------------------------------------------------------------------- /figures/exploiting_llm_quantization.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/exploiting_llm_quantization.png -------------------------------------------------------------------------------- /figures/fast-2-bit.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/fast-2-bit.png -------------------------------------------------------------------------------- /figures/flashattention3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/flashattention3.png -------------------------------------------------------------------------------- /figures/high_sparsity_pretraining.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/high_sparsity_pretraining.png -------------------------------------------------------------------------------- /figures/homer.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/homer.png -------------------------------------------------------------------------------- /figures/hybridLLM.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/hybridLLM.png -------------------------------------------------------------------------------- /figures/hymba.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/hymba.png -------------------------------------------------------------------------------- /figures/impossible_distillation.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/impossible_distillation.png -------------------------------------------------------------------------------- /figures/junk_DNA.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/junk_DNA.png -------------------------------------------------------------------------------- /figures/kcache.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/kcache.png -------------------------------------------------------------------------------- /figures/kd_close_source.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/kd_close_source.png -------------------------------------------------------------------------------- /figures/kvzip.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/kvzip.png -------------------------------------------------------------------------------- /figures/learn-to-reason.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/learn-to-reason.png -------------------------------------------------------------------------------- /figures/lillama.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/lillama.png -------------------------------------------------------------------------------- /figures/llm_pruner.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/llm_pruner.png -------------------------------------------------------------------------------- /figures/llma.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/llma.png -------------------------------------------------------------------------------- /figures/longctx_bench.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/longctx_bench.png -------------------------------------------------------------------------------- /figures/longllmlingua.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/longllmlingua.png -------------------------------------------------------------------------------- /figures/mamba_drafters.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/mamba_drafters.png -------------------------------------------------------------------------------- /figures/megalodon.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/megalodon.png -------------------------------------------------------------------------------- /figures/minicache.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/minicache.png -------------------------------------------------------------------------------- /figures/minillm.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/minillm.png -------------------------------------------------------------------------------- /figures/mixtral_offloading.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/mixtral_offloading.png -------------------------------------------------------------------------------- /figures/multiflow.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/multiflow.png -------------------------------------------------------------------------------- /figures/neuroal.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/neuroal.png -------------------------------------------------------------------------------- /figures/nugget2D.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/nugget2D.png -------------------------------------------------------------------------------- /figures/omniquant.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/omniquant.png -------------------------------------------------------------------------------- /figures/outliersuppression.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/outliersuppression.png -------------------------------------------------------------------------------- /figures/parrot.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/parrot.png -------------------------------------------------------------------------------- /figures/preble.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/preble.png -------------------------------------------------------------------------------- /figures/pv-tuning.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/pv-tuning.png -------------------------------------------------------------------------------- /figures/qllm-eval.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/qllm-eval.png -------------------------------------------------------------------------------- /figures/qlora.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/qlora.png -------------------------------------------------------------------------------- /figures/quantized-lm-confidence.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/quantized-lm-confidence.png -------------------------------------------------------------------------------- /figures/recall_and_icl.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/recall_and_icl.png -------------------------------------------------------------------------------- /figures/reform.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/reform.png -------------------------------------------------------------------------------- /figures/relu2wins.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/relu2wins.png -------------------------------------------------------------------------------- /figures/relufication.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/relufication.png -------------------------------------------------------------------------------- /figures/rethink-AKL.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/rethink-AKL.png -------------------------------------------------------------------------------- /figures/selective-cloud-llm-assistance.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/selective-cloud-llm-assistance.png -------------------------------------------------------------------------------- /figures/selective_context.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/selective_context.png -------------------------------------------------------------------------------- /figures/sensitivity_sparse.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/sensitivity_sparse.png -------------------------------------------------------------------------------- /figures/simba.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/simba.png -------------------------------------------------------------------------------- /figures/sparsegpt.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/sparsegpt.png -------------------------------------------------------------------------------- /figures/speccascade.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/speccascade.png -------------------------------------------------------------------------------- /figures/spinquant.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/spinquant.png -------------------------------------------------------------------------------- /figures/stand.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/stand.png -------------------------------------------------------------------------------- /figures/survey_xunyu.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/survey_xunyu.png -------------------------------------------------------------------------------- /figures/switch.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/switch.png -------------------------------------------------------------------------------- /figures/switchhead.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/switchhead.png -------------------------------------------------------------------------------- /figures/unrolled.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/unrolled.png -------------------------------------------------------------------------------- /figures/watermark_quant.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/watermark_quant.png -------------------------------------------------------------------------------- /figures/zephyr.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/zephyr.png -------------------------------------------------------------------------------- /figures/zeroquant-6bit.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/zeroquant-6bit.png -------------------------------------------------------------------------------- /figures/zeroquant-v2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/zeroquant-v2.png -------------------------------------------------------------------------------- /figures/zipcache.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/figures/zipcache.png -------------------------------------------------------------------------------- /generate_item.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/generate_item.py -------------------------------------------------------------------------------- /hardware.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/hardware.md -------------------------------------------------------------------------------- /inference_acceleration.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/inference_acceleration.md -------------------------------------------------------------------------------- /knowledge_distillation.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/knowledge_distillation.md -------------------------------------------------------------------------------- /kv_cache_compression.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/kv_cache_compression.md -------------------------------------------------------------------------------- /leaderboard.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/leaderboard.md -------------------------------------------------------------------------------- /low_rank_decomposition.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/low_rank_decomposition.md -------------------------------------------------------------------------------- /project/readme.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/project/readme.md -------------------------------------------------------------------------------- /pruning.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/pruning.md -------------------------------------------------------------------------------- /quantization.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/quantization.md -------------------------------------------------------------------------------- /survey.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/survey.md -------------------------------------------------------------------------------- /text_compression.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/text_compression.md -------------------------------------------------------------------------------- /tuning.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/horseee/Awesome-Efficient-LLM/HEAD/tuning.md --------------------------------------------------------------------------------