| .. |
|
__init__.py
|
7ca63930c8
support deepseek_v3 model
|
vor 4 Wochen |
|
arctic.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
baichuan.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
bart.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
blip.py
|
f56d6b396a
vlm: fallback to SDPA for ViT models on CPU backend (#982)
|
vor 1 Monat |
|
blip2.py
|
b4a1e2fd02
vlm: add tensor parallel support for vision transformer models (#971)
|
vor 1 Monat |
|
bloom.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
chameleon.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
chatglm.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
clip.py
|
4d14bd1fe5
vlm: add multi-input support for LLaVA and InternVL models (#1002)
|
vor 1 Monat |
|
commandr.py
|
135dfd648b
fix: LoRA support for Cohere and Jamba models (#1004)
|
vor 1 Monat |
|
dbrx.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
decilm.py
|
9022c6d869
remove progress_bar imports
|
vor 3 Monaten |
|
deepseek.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
deepseek_v2.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
deepseek_v3.py
|
7ca63930c8
support deepseek_v3 model
|
vor 4 Wochen |
|
eagle.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
exaone.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
falcon.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
fuyu.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
gemma.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
gemma2.py
|
b33cf04386
quants: add bitsandbytes support for gemma2 model (#1026)
|
vor 4 Wochen |
|
gpt2.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
gpt_bigcode.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
gpt_j.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
gpt_neox.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
granite.py
|
411ac4f405
vlm: add support for Qwen2-VL model (#1015)
|
vor 1 Monat |
|
idefics2_vision_model.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
vor 4 Monaten |
|
interfaces.py
|
0b8b407b6d
feat: support profiling with multiple multi-modal inputs per prompt (#712)
|
vor 4 Monaten |
|
intern_vit.py
|
f56d6b396a
vlm: fallback to SDPA for ViT models on CPU backend (#982)
|
vor 1 Monat |
|
internlm2.py
|
7632f91429
fix: InternLM2 model with Tensor Parallel (#980)
|
vor 1 Monat |
|
internvl.py
|
41ceb754a6
vlm: fix internvl2 inference with various num_patches (#1030)
|
vor 4 Wochen |
|
jais.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
jamba.py
|
135dfd648b
fix: LoRA support for Cohere and Jamba models (#1004)
|
vor 1 Monat |
|
llama.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
llama_embedding.py
|
9022c6d869
remove progress_bar imports
|
vor 3 Monaten |
|
llava.py
|
4d14bd1fe5
vlm: add multi-input support for LLaVA and InternVL models (#1002)
|
vor 1 Monat |
|
llava_next.py
|
4d14bd1fe5
vlm: add multi-input support for LLaVA and InternVL models (#1002)
|
vor 1 Monat |
|
llava_next_video.py
|
be59e30139
vlm: add support for video modality + llava next video (#1014)
|
vor 1 Monat |
|
mamba.py
|
3bb0f07461
chore: rename `task_handler` to `worker` (#985)
|
vor 1 Monat |
|
mamba_cache.py
|
a113309876
kernel: add meta functions for ops to prevent graph breaks (#1019)
|
vor 4 Wochen |
|
medusa.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
minicpm.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
minicpmv.py
|
548e864404
models: add support for QwenVL (#995)
|
vor 1 Monat |
|
mixtral.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
mixtral_quant.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
mlp_speculator.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
mpt.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
na_vit.py
|
9f3e7c86e2
feat: add fused Marlin MoE kernel (#934)
|
vor 1 Monat |
|
nemotron.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
olmo.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
olmoe.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
opt.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
orion.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
paligemma.py
|
46d577f019
vlm: fix siglip layernorm and paligemma weight loading (#991)
|
vor 1 Monat |
|
persimmon.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
phi.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
phi3_small.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
phi3v.py
|
4d14bd1fe5
vlm: add multi-input support for LLaVA and InternVL models (#1002)
|
vor 1 Monat |
|
phimoe.py
|
201db10f02
models: add support for Phi3 MoE
|
vor 1 Monat |
|
pixtral.py
|
1721bea53a
vlm: add support for Pixtral model (#1022)
|
vor 4 Wochen |
|
qwen.py
|
8d5d87e687
vlm: support multiple images for qwen-vl (#1031)
|
vor 4 Wochen |
|
qwen2.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
qwen2_moe.py
|
5224389dae
chore: skip loading extra bias for qwen2 moe GPTQ (#1011)
|
vor 1 Monat |
|
qwen2_vl.py
|
411ac4f405
vlm: add support for Qwen2-VL model (#1015)
|
vor 1 Monat |
|
siglip.py
|
4d14bd1fe5
vlm: add multi-input support for LLaVA and InternVL models (#1002)
|
vor 1 Monat |
|
solar.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
stablelm.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
starcoder2.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
|
ultravox.py
|
8eb4a3cfd3
vlm: support multiple audios per prompt for Ultravox (#990)
|
vor 1 Monat |
|
utils.py
|
a8bdd488b9
distributed: support pipeline parallelism for internvl and internlm2 (#965)
|
vor 1 Monat |
|
xverse.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 1 Monat |
AlpinDale
7ca63930c8
support deepseek_v3 model