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