 AlpinDale
|
22429990b2
alias sampler_priority to sampler_order
|
hace 1 mes |
|
AlpinDale
|
224249e608
Merge branch 'main' into sampler_priorty
|
hace 1 mes |
 AlpinDale
|
93bc863591
feat: Machete Kernels for Hopper GPUs (#842)
|
hace 1 mes |
|
AlpinDale
|
563e8f7ac8
fix: latency and serving benchmarks (#841)
|
hace 1 mes |
|
AlpinDale
|
7c7ec12f36
chore: refactor executor classes for easier inheritance (#840)
|
hace 1 mes |
|
AlpinDale
|
16b587c104
fix: hidden states handling in batch expansion for spec decoding (#839)
|
hace 1 mes |
|
AlpinDale
|
224284d232
more cleaning up and logs
|
hace 1 mes |
|
AlpinDale
|
2242c3e098
more clean-up and remove min_tokens from the order
|
hace 1 mes |
|
AlpinDale
|
22427746c6
fix: sampler arg verification
|
hace 1 mes |
|
AlpinDale
|
2242ad3ecf
feat: add sampler_priorty
|
hace 1 mes |
|
AlpinDale
|
60f7b828d5
feat: add skew sampling (#834)
|
hace 2 meses |
|
AlpinDale
|
ba9d8f631a
feat: add no_repeat_ngram sampler (#832)
|
hace 2 meses |
 Selali
|
4c4a365f77
feat: Add DRY (Don't Repeat Yourself) sampling (#827)
|
hace 2 meses |
|
AlpinDale
|
48a8693aed
feat: multi-step scheduling (#831)
|
hace 2 meses |
|
AlpinDale
|
2242cb25dc
fix: unbound tokenizer error
|
hace 2 meses |
|
AlpinDale
|
3d83e64f8e
feat: add metrics for prefix cache hit rate (#829)
|
hace 2 meses |
|
AlpinDale
|
22425b689d
fix: XPU build
|
hace 2 meses |
|
AlpinDale
|
bfc8988116
feat: add cuda sampling kernels for top_k and top_p (#828)
|
hace 2 meses |
|
AlpinDale
|
22427602eb
feat: add top-nsigma sampling method
|
hace 2 meses |
|
AlpinDale
|
22429e4a10
fix: sampler test with new transformers version
|
hace 2 meses |
|
AlpinDale
|
2f61644f6e
SPMD optimizations (#824)
|
hace 2 meses |
|
AlpinDale
|
32a37e8107
tests: partially fix tensorizer and logprobs tests
|
hace 2 meses |
|
AlpinDale
|
7f1c9af5e2
fix: fp8 quant test
|
hace 2 meses |
|
AlpinDale
|
173ac23399
fix: experts int8 quant test
|
hace 2 meses |
|
AlpinDale
|
68f050129d
fix: lora worker manager test import
|
hace 2 meses |
|
AlpinDale
|
3661de812d
fix: lora layer test
|
hace 2 meses |
|
AlpinDale
|
0a369f9171
feat: support chunked prefill with LoRA (#823)
|
hace 2 meses |
|
AlpinDale
|
e5b1afe625
feat: add chat method for LLM class (#822)
|
hace 2 meses |
|
AlpinDale
|
262cbc63b7
fix: vision api test template path
|
hace 2 meses |
|
AlpinDale
|
b0113a1eaa
fix: tokenization api test (#821)
|
hace 2 meses |
