PrototypeCommon
fkodom/grouped-query-attention-pytorch
(Unofficial) PyTorch implementation of grouped-query attention (GQA) from "GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints" (https://arxiv.org/pdf/2305.13245.pdf)
- Owner
- fkodom (personal account)
- Stars
- Stars: 196
- Forks
- Forks: 14
- Language
- Python
- License
- MIT
- Created
- 2023
- Tier
- Prototype (100 to 249 ★) What this tier means
More repos by fkodom
All 4 repos by fkodom- fkodom/fft-conv-pytorchStars: 524PythonAlpha
- fkodom/clip-text-decoderStars: 119PythonPrototype
- fkodom/yet-another-retnetStars: 105PythonPrototype
Nearby in Python
All Python repos- knoic/ComfyUI-MiniMaxH3-PrefixStreamStars: 196PythonPrototype
- kusaida/AdwairStars: 196PythonPrototype
- sepehr071/iran-mcpStars: 196PythonPrototype
- patpadgett/corkscrewStars: 196PythonPrototype
- wbond/sublime_prefixrStars: 196PythonPrototype
- seomoz/reppyStars: 196PythonPrototype