├── .devops ├── full.Dockerfile ├── main.Dockerfile └── tools.sh ├── .dockerignore ├── .ecrc ├── .editorconfig ├── .github ├── ISSUE_TEMPLATE │ └── custom.md └── workflows │ ├── build.yml │ ├── docker.yml │ └── editorconfig.yml ├── .gitignore ├── CMakeLists.txt ├── LICENSE ├── Makefile ├── Package.swift ├── README.md ├── SHA256SUMS ├── convert-pth-to-ggml.py ├── convert.py ├── examples ├── CMakeLists.txt ├── Miku.sh ├── alpaca.sh ├── benchmark │ └── benchmark-q4_0-matmult.c ├── chat-13B.bat ├── chat-13B.sh ├── chat.sh ├── common.cpp ├── common.h ├── embedding │ ├── CMakeLists.txt │ ├── README.md │ └── embedding.cpp ├── embeddings-benchmark │ └── embeddings-benchmark.cpp ├── embeddings-server-client.py ├── embeddings-server │ └── embeddings-server.cpp ├── gpt4all.sh ├── main │ ├── CMakeLists.txt │ ├── README.md │ └── main.cpp ├── mteb-benchmark.py ├── perplexity │ ├── CMakeLists.txt │ ├── README.md │ └── perplexity.cpp ├── quantize-stats │ ├── CMakeLists.txt │ └── quantize-stats.cpp ├── quantize │ ├── CMakeLists.txt │ ├── README.md │ └── quantize.cpp └── reason-act.sh ├── flake.lock ├── flake.nix ├── ggml.c ├── ggml.h ├── llama.cpp ├── llama.h ├── llama_util.h ├── media ├── llama-leader.jpeg ├── llama0-banner.png ├── llama0-logo.png ├── llama1-banner.png └── llama1-logo.png ├── model_creation ├── download_q4_weights.py ├── ggml-conversion.sh ├── model.py └── upload_to_hf.py ├── models └── ggml-vocab.bin ├── mteb-results-7B └── BIOSSES.json ├── mteb-results ├── BIOSSES.json ├── STS12.json ├── STS16.json └── STSBenchmark.json ├── prompts ├── alpaca.txt ├── chat-with-bob.txt ├── dan.txt └── reason-act.txt ├── requirements.txt ├── spm-headers └── llama.h └── tests ├── CMakeLists.txt ├── test-double-float.c ├── test-quantize.c └── test-tokenizer-0.cpp /.devops/full.Dockerfile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.devops/full.Dockerfile -------------------------------------------------------------------------------- /.devops/main.Dockerfile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.devops/main.Dockerfile -------------------------------------------------------------------------------- /.devops/tools.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.devops/tools.sh -------------------------------------------------------------------------------- /.dockerignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.dockerignore -------------------------------------------------------------------------------- /.ecrc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.ecrc -------------------------------------------------------------------------------- /.editorconfig: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.editorconfig -------------------------------------------------------------------------------- /.github/ISSUE_TEMPLATE/custom.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.github/ISSUE_TEMPLATE/custom.md -------------------------------------------------------------------------------- /.github/workflows/build.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.github/workflows/build.yml -------------------------------------------------------------------------------- /.github/workflows/docker.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.github/workflows/docker.yml -------------------------------------------------------------------------------- /.github/workflows/editorconfig.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.github/workflows/editorconfig.yml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/.gitignore -------------------------------------------------------------------------------- /CMakeLists.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/CMakeLists.txt -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/LICENSE -------------------------------------------------------------------------------- /Makefile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/Makefile -------------------------------------------------------------------------------- /Package.swift: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/Package.swift -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/README.md -------------------------------------------------------------------------------- /SHA256SUMS: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/SHA256SUMS -------------------------------------------------------------------------------- /convert-pth-to-ggml.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/convert-pth-to-ggml.py -------------------------------------------------------------------------------- /convert.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/convert.py -------------------------------------------------------------------------------- /examples/CMakeLists.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/CMakeLists.txt -------------------------------------------------------------------------------- /examples/Miku.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/Miku.sh -------------------------------------------------------------------------------- /examples/alpaca.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/alpaca.sh -------------------------------------------------------------------------------- /examples/benchmark/benchmark-q4_0-matmult.c: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/benchmark/benchmark-q4_0-matmult.c -------------------------------------------------------------------------------- /examples/chat-13B.bat: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/chat-13B.bat -------------------------------------------------------------------------------- /examples/chat-13B.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/chat-13B.sh -------------------------------------------------------------------------------- /examples/chat.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/chat.sh -------------------------------------------------------------------------------- /examples/common.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/common.cpp -------------------------------------------------------------------------------- /examples/common.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/common.h -------------------------------------------------------------------------------- /examples/embedding/CMakeLists.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/embedding/CMakeLists.txt -------------------------------------------------------------------------------- /examples/embedding/README.md: -------------------------------------------------------------------------------- 1 | # embedding 2 | 3 | TODO 4 | -------------------------------------------------------------------------------- /examples/embedding/embedding.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/embedding/embedding.cpp -------------------------------------------------------------------------------- /examples/embeddings-benchmark/embeddings-benchmark.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/embeddings-benchmark/embeddings-benchmark.cpp -------------------------------------------------------------------------------- /examples/embeddings-server-client.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/embeddings-server-client.py -------------------------------------------------------------------------------- /examples/embeddings-server/embeddings-server.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/embeddings-server/embeddings-server.cpp -------------------------------------------------------------------------------- /examples/gpt4all.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/gpt4all.sh -------------------------------------------------------------------------------- /examples/main/CMakeLists.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/main/CMakeLists.txt -------------------------------------------------------------------------------- /examples/main/README.md: -------------------------------------------------------------------------------- 1 | # main 2 | 3 | TODO 4 | -------------------------------------------------------------------------------- /examples/main/main.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/main/main.cpp -------------------------------------------------------------------------------- /examples/mteb-benchmark.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/mteb-benchmark.py -------------------------------------------------------------------------------- /examples/perplexity/CMakeLists.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/perplexity/CMakeLists.txt -------------------------------------------------------------------------------- /examples/perplexity/README.md: -------------------------------------------------------------------------------- 1 | # perplexity 2 | 3 | TODO 4 | -------------------------------------------------------------------------------- /examples/perplexity/perplexity.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/perplexity/perplexity.cpp -------------------------------------------------------------------------------- /examples/quantize-stats/CMakeLists.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/quantize-stats/CMakeLists.txt -------------------------------------------------------------------------------- /examples/quantize-stats/quantize-stats.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/quantize-stats/quantize-stats.cpp -------------------------------------------------------------------------------- /examples/quantize/CMakeLists.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/quantize/CMakeLists.txt -------------------------------------------------------------------------------- /examples/quantize/README.md: -------------------------------------------------------------------------------- 1 | # quantize 2 | 3 | TODO 4 | -------------------------------------------------------------------------------- /examples/quantize/quantize.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/quantize/quantize.cpp -------------------------------------------------------------------------------- /examples/reason-act.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/examples/reason-act.sh -------------------------------------------------------------------------------- /flake.lock: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/flake.lock -------------------------------------------------------------------------------- /flake.nix: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/flake.nix -------------------------------------------------------------------------------- /ggml.c: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/ggml.c -------------------------------------------------------------------------------- /ggml.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/ggml.h -------------------------------------------------------------------------------- /llama.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/llama.cpp -------------------------------------------------------------------------------- /llama.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/llama.h -------------------------------------------------------------------------------- /llama_util.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/llama_util.h -------------------------------------------------------------------------------- /media/llama-leader.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/media/llama-leader.jpeg -------------------------------------------------------------------------------- /media/llama0-banner.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/media/llama0-banner.png -------------------------------------------------------------------------------- /media/llama0-logo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/media/llama0-logo.png -------------------------------------------------------------------------------- /media/llama1-banner.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/media/llama1-banner.png -------------------------------------------------------------------------------- /media/llama1-logo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/media/llama1-logo.png -------------------------------------------------------------------------------- /model_creation/download_q4_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/model_creation/download_q4_weights.py -------------------------------------------------------------------------------- /model_creation/ggml-conversion.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/model_creation/ggml-conversion.sh -------------------------------------------------------------------------------- /model_creation/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/model_creation/model.py -------------------------------------------------------------------------------- /model_creation/upload_to_hf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/model_creation/upload_to_hf.py -------------------------------------------------------------------------------- /models/ggml-vocab.bin: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/models/ggml-vocab.bin -------------------------------------------------------------------------------- /mteb-results-7B/BIOSSES.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/mteb-results-7B/BIOSSES.json -------------------------------------------------------------------------------- /mteb-results/BIOSSES.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/mteb-results/BIOSSES.json -------------------------------------------------------------------------------- /mteb-results/STS12.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/mteb-results/STS12.json -------------------------------------------------------------------------------- /mteb-results/STS16.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/mteb-results/STS16.json -------------------------------------------------------------------------------- /mteb-results/STSBenchmark.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/mteb-results/STSBenchmark.json -------------------------------------------------------------------------------- /prompts/alpaca.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/prompts/alpaca.txt -------------------------------------------------------------------------------- /prompts/chat-with-bob.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/prompts/chat-with-bob.txt -------------------------------------------------------------------------------- /prompts/dan.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/prompts/dan.txt -------------------------------------------------------------------------------- /prompts/reason-act.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/prompts/reason-act.txt -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/requirements.txt -------------------------------------------------------------------------------- /spm-headers/llama.h: -------------------------------------------------------------------------------- 1 | ../llama.h -------------------------------------------------------------------------------- /tests/CMakeLists.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/tests/CMakeLists.txt -------------------------------------------------------------------------------- /tests/test-double-float.c: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/tests/test-double-float.c -------------------------------------------------------------------------------- /tests/test-quantize.c: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/tests/test-quantize.c -------------------------------------------------------------------------------- /tests/test-tokenizer-0.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/skeskinen/llama-lite/HEAD/tests/test-tokenizer-0.cpp --------------------------------------------------------------------------------