 AlpinDale
|
fb4c01740c
feat: add asymmetric TP support for Qwen2
|
hai 5 meses |
|
AlpinDale
|
9d7beaa5b9
chore: separate kv_scale into k_scale and v_scale
|
hai 5 meses |
 AlpinDale
|
0f4a9ee77b
quantized lm_head (#582)
|
hai 5 meses |
|
AlpinDale
|
ae04f57ec1
feat: Pipeline Parallel support (#581)
|
hai 6 meses |
|
AlpinDale
|
c5d8028668
fix: no need to redefine supports_vision and supports_lora in model class
|
hai 6 meses |
|
AlpinDale
|
56e0b8223c
chore: add base class for LoRA-supported models
|
hai 6 meses |
|
AlpinDale
|
025322ee5f
fix: fp8 kv cache for qwen2 models
|
hai 6 meses |
|
AlpinDale
|
ac79d115b3
add guards for prefix caching, fp8, chunked, etc
|
hai 6 meses |
|
AlpinDale
|
656459fd84
make fp8_e4m3 work on nvidia
|
hai 6 meses |
|
AlpinDale
|
295cfb2f39
add rope scaling for qwen2
|
hai 6 meses |
|
AlpinDale
|
50b7c13db0
refactor: attention selector (#552)
|
hai 6 meses |
|
AlpinDale
|
b178ae4b4a
chore: generalize linear_method to be quant_method (#540)
|
hai 6 meses |
|
AlpinDale
|
fca911ee0a
vLLM Upstream Sync (#526)
|
hai 7 meses |
|
AlpinDale
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
hai 9 meses |
|
AlpinDale
|
f8dfac6372
chore: attention refactor and upstream sync apr01 (#365)
|
hai 10 meses |
|
AlpinDale
|
da223153c6
feat&fix: cohere support and missing GPU blocks (#333)
|
hai 11 meses |
|
AlpinDale
|
e42a78381a
feat: switch from pylint to ruff (#322)
|
hai 11 meses |
|
AlpinDale
|
e31c6f0b45
feat: refactor modeling logic and support more models (#274)
|
hai 11 meses |
