| .. |
|
quantization
|
f8652c8e99
fix: optimize aqlm dequantization (#325)
|
пре 10 месеци |
|
triton_kernel
|
e42a78381a
feat: switch from pylint to ruff (#322)
|
пре 10 месеци |
|
__init__.py
|
07aa2a492f
upstream: add option to specify tokenizer
|
пре 1 година |
|
activation.py
|
e31c6f0b45
feat: refactor modeling logic and support more models (#274)
|
пре 11 месеци |
|
attention.py
|
9810daa699
feat: INT8 KV Cache (#298)
|
пре 10 месеци |
|
layernorm.py
|
e31c6f0b45
feat: refactor modeling logic and support more models (#274)
|
пре 11 месеци |
|
linear.py
|
e42a78381a
feat: switch from pylint to ruff (#322)
|
пре 10 месеци |
|
rejection.py
|
95bdd35ec9
feat: rejection sampler (#197)
|
пре 1 година |
|
rotary_embedding.py
|
e42a78381a
feat: switch from pylint to ruff (#322)
|
пре 10 месеци |
|
sampler.py
|
da223153c6
feat&fix: cohere support and missing GPU blocks (#333)
|
пре 10 месеци |
|
vocab_parallel_embedding.py
|
968bde81bf
fix: tensor parallel with GPTQ and AWQ quants (#307)
|
пре 10 месеци |
AlpinDale
da223153c6
feat&fix: cohere support and missing GPU blocks (#333)