├── LICENSE ├── Readme.md ├── assets ├── logo.png └── swift.png ├── eval_llama.sh ├── evaluation_llama ├── eval.py ├── inference_baseline.py ├── inference_swift.py └── speed.py ├── model └── swift │ ├── kv_cache.py │ ├── modeling_llama.py │ └── utils.py ├── requirements.txt └── skip_layers.json /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/LICENSE -------------------------------------------------------------------------------- /Readme.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/Readme.md -------------------------------------------------------------------------------- /assets/logo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/assets/logo.png -------------------------------------------------------------------------------- /assets/swift.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/assets/swift.png -------------------------------------------------------------------------------- /eval_llama.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/eval_llama.sh -------------------------------------------------------------------------------- /evaluation_llama/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/evaluation_llama/eval.py -------------------------------------------------------------------------------- /evaluation_llama/inference_baseline.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/evaluation_llama/inference_baseline.py -------------------------------------------------------------------------------- /evaluation_llama/inference_swift.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/evaluation_llama/inference_swift.py -------------------------------------------------------------------------------- /evaluation_llama/speed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/evaluation_llama/speed.py -------------------------------------------------------------------------------- /model/swift/kv_cache.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/model/swift/kv_cache.py -------------------------------------------------------------------------------- /model/swift/modeling_llama.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/model/swift/modeling_llama.py -------------------------------------------------------------------------------- /model/swift/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/model/swift/utils.py -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/requirements.txt -------------------------------------------------------------------------------- /skip_layers.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hemingkx/SWIFT/HEAD/skip_layers.json --------------------------------------------------------------------------------