Strona główna Odkrywaj Pomoc
Zarejestruj się Zaloguj się
david
/
aphrodite-engine
kopia lustrzana https://github.com/PygmalionAI/aphrodite-engine
1
0
Forkuj 0
Pliki Problemy 0 Wiki
Drzewo: 172dee2573
Gałęzie Tagi
aphrodite-engin...
/
aphrodite
/
modeling
/
models
AlpinDale fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 1 miesiąc temu
..
__init__.py a5bfc2bc3d VLM: add support for LLaVA-Onevision model (#1100) 1 miesiąc temu
arctic.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
baichuan.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
bart.py a985143768 core: add cuda graph support for encoder-decoder models (#1051) 2 miesięcy temu
blip.py 7ce3174039 VLM: refactor blip models to support composite weight loading (#1098) 1 miesiąc temu
blip2.py 7ce3174039 VLM: refactor blip models to support composite weight loading (#1098) 1 miesiąc temu
bloom.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
chameleon.py 651678d2df VLM: use `SequenceData.from_token_counts` to create dummy data (#1093) 1 miesiąc temu
chatglm.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
clip.py 80be38ca6f chore: expose phi3_v num_crops as an mm_processor_kwargs (#1117) 1 miesiąc temu
commandr.py 135dfd648b fix: LoRA support for Cohere and Jamba models (#1004) 2 miesięcy temu
dbrx.py b65449b5ad moe: refactor DBRX experts to support FusedMoE (#1095) 1 miesiąc temu
decilm.py 9022c6d869 remove progress_bar imports 4 miesięcy temu
deepseek.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
deepseek_v2.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
eagle.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
exaone.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
falcon.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
fuyu.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
gemma.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
gemma2.py b33cf04386 quants: add bitsandbytes support for gemma2 model (#1026) 2 miesięcy temu
gpt2.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
gpt_bigcode.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
gpt_j.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
gpt_neox.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
granite.py 11f49b5341 fix: granite logit scale in logit computation (#1054) 2 miesięcy temu
idefics2_vision_model.py f1d0b77c92 [0.6.0] Release Candidate (#481) 6 miesięcy temu
interfaces.py 0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712) 6 miesięcy temu
intern_vit.py f56d6b396a vlm: fallback to SDPA for ViT models on CPU backend (#982) 2 miesięcy temu
internlm2.py 7632f91429 fix: InternLM2 model with Tensor Parallel (#980) 2 miesięcy temu
internvl.py 91d03c04d2 VLM: refactor composite weight loading logic (#1097) 1 miesiąc temu
jais.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
jamba.py 135dfd648b fix: LoRA support for Cohere and Jamba models (#1004) 2 miesięcy temu
llama.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
llama_embedding.py 9022c6d869 remove progress_bar imports 4 miesięcy temu
llava.py 91d03c04d2 VLM: refactor composite weight loading logic (#1097) 1 miesiąc temu
llava_next.py 7ce3174039 VLM: refactor blip models to support composite weight loading (#1098) 1 miesiąc temu
llava_next_video.py 91d03c04d2 VLM: refactor composite weight loading logic (#1097) 1 miesiąc temu
llava_onevision.py a5bfc2bc3d VLM: add support for LLaVA-Onevision model (#1100) 1 miesiąc temu
mamba.py 3bb0f07461 chore: rename `task_handler` to `worker` (#985) 2 miesięcy temu
mamba_cache.py a113309876 kernel: add meta functions for ops to prevent graph breaks (#1019) 2 miesięcy temu
medusa.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
minicpm.py ce7b602f03 model: add support for MiniCPM-3 (#1044) 2 miesięcy temu
minicpm3.py ce7b602f03 model: add support for MiniCPM-3 (#1044) 2 miesięcy temu
minicpmv.py 651678d2df VLM: use `SequenceData.from_token_counts` to create dummy data (#1093) 1 miesiąc temu
mixtral.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
mixtral_quant.py fa84f8102e kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119) 1 miesiąc temu
mlp_speculator.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
molmo.py acc0c727c8 vlm: add support for molmo vision model (#1069) 2 miesięcy temu
mpt.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
na_vit.py 9f3e7c86e2 feat: add fused Marlin MoE kernel (#934) 2 miesięcy temu
nemotron.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
olmo.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
olmoe.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
opt.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
orion.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
paligemma.py 91d03c04d2 VLM: refactor composite weight loading logic (#1097) 1 miesiąc temu
persimmon.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
phi.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
phi3.py ec17b6c4d0 fix: Phi3.5 Mini and MoE LoRA inference (#1070) 2 miesięcy temu
phi3_small.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
phi3v.py 80be38ca6f chore: expose phi3_v num_crops as an mm_processor_kwargs (#1117) 1 miesiąc temu
phimoe.py ec17b6c4d0 fix: Phi3.5 Mini and MoE LoRA inference (#1070) 2 miesięcy temu
pixtral.py 651678d2df VLM: use `SequenceData.from_token_counts` to create dummy data (#1093) 1 miesiąc temu
qwen.py 651678d2df VLM: use `SequenceData.from_token_counts` to create dummy data (#1093) 1 miesiąc temu
qwen2.py c90abcc603 VLM: add pipeline parallelism support for Qwen2-VL (#1103) 1 miesiąc temu
qwen2_moe.py 5224389dae chore: skip loading extra bias for qwen2 moe GPTQ (#1011) 2 miesięcy temu
qwen2_vl.py c90abcc603 VLM: add pipeline parallelism support for Qwen2-VL (#1103) 1 miesiąc temu
siglip.py a5bfc2bc3d VLM: add support for LLaVA-Onevision model (#1100) 1 miesiąc temu
solar.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
stablelm.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
starcoder2.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu
ultravox.py 18e0b0e932 chore: support loading weights by ID within models (#1116) 1 miesiąc temu
utils.py 91d03c04d2 VLM: refactor composite weight loading logic (#1097) 1 miesiąc temu
xverse.py 0dfa6b60ec core: support logprobs with multi-step scheduling (#963) 2 miesięcy temu