david/aphrodite-engine: PygmalionAI's large-scale inference engine pygmalion.chat It is designed to serve as the inference endpoint for the PygmalionAI website, and to allow serving the Pygmalion models to a large number of users with blazing fast speeds (thanks to vLLM's Paged Attention). @ dry

Geun, Lim 08711d2ac9 feat: add Exaone model support (#743)		3 месяцев назад
..
__init__.py	08711d2ac9 feat: add Exaone model support (#743)	3 месяцев назад
arctic.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
baichuan.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
bart.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
blip.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
blip2.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
bloom.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
chameleon.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
chatglm.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
clip.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
commandr.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
dbrx.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
decilm.py	3f712cd287 feat: add progress bar for loading individual weight modules (#640)	4 месяцев назад
deepseek.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
deepseek_v2.py	4ec08af18b chore: update fused MoE weight loading (#700)	4 месяцев назад
exaone.py	08711d2ac9 feat: add Exaone model support (#743)	3 месяцев назад
falcon.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
fuyu.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
gemma.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
gemma2.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
gpt2.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
gpt_bigcode.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
gpt_j.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
gpt_neox.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
idefics2_vision_model.py	f1d0b77c92 [0.6.0] Release Candidate (#481)	4 месяцев назад
interfaces.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
intern_vit.py	3f712cd287 feat: add progress bar for loading individual weight modules (#640)	4 месяцев назад
internlm2.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
internvl.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
jais.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
jamba.py	d34e083c48 feat: add experts_int8 support (#730)	3 месяцев назад
llama.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
llama_embedding.py	3f712cd287 feat: add progress bar for loading individual weight modules (#640)	4 месяцев назад
llava.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
llava_next.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
mamba.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
mamba_cache.py	bf88c8567e feat: mamba model support (#674)	4 месяцев назад
medusa.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
minicpm.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
minicpmv.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
mixtral.py	4ec08af18b chore: update fused MoE weight loading (#700)	4 месяцев назад
mixtral_quant.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
mlp_speculator.py	3f712cd287 feat: add progress bar for loading individual weight modules (#640)	4 месяцев назад
mpt.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
na_vit.py	f1d0b77c92 [0.6.0] Release Candidate (#481)	4 месяцев назад
nemotron.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
olmo.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
opt.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
orion.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
paligemma.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
persimmon.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
phi.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
phi3_small.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
phi3v.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
qwen.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
qwen2.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
qwen2_moe.py	4ec08af18b chore: update fused MoE weight loading (#700)	4 месяцев назад
siglip.py	0b8b407b6d feat: support profiling with multiple multi-modal inputs per prompt (#712)	4 месяцев назад
solar.py	0c6f03b7e4 feat: add Solar model support (#701)	4 месяцев назад
stablelm.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
starcoder2.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад
utils.py	3693028340 feat: support for Audio modality (#698)	4 месяцев назад
xverse.py	0e558e9b2f fix: loading chameleon model with TP>1 (#695)	4 месяцев назад