A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 17, 08:00 UTC
Update Supported Model List (#825)
Zhuohan Li committed
Aug 22, 2023 at 11:51 UTC
14f9c72bfdd6c90cbcb52bcfa34f33f56cbe323f
2 files changed
+12
README.md
+3
@@ -42,6 +42,7 @@ vLLM is flexible and easy to use with:
42
43
vLLM seamlessly supports many Huggingface models, including the following architectures:
44
45
+- Aqualia (`BAAI/Aquila-7B`, `BAAI/AquilaChat-7B`, etc.)
46
- Baichuan (`baichuan-inc/Baichuan-7B`, `baichuan-inc/Baichuan-13B-Chat`, etc.)
47
- BLOOM (`bigscience/bloom`, `bigscience/bloomz`, etc.)
48
- Falcon (`tiiuae/falcon-7b`, `tiiuae/falcon-40b`, `tiiuae/falcon-rw-7b`, etc.)
@@ -49,9 +50,11 @@ vLLM seamlessly supports many Huggingface models, including the following archit
50
- GPT BigCode (`bigcode/starcoder`, `bigcode/gpt_bigcode-santacoder`, etc.)
51
- GPT-J (`EleutherAI/gpt-j-6b`, `nomic-ai/gpt4all-j`, etc.)
52
- GPT-NeoX (`EleutherAI/gpt-neox-20b`, `databricks/dolly-v2-12b`, `stabilityai/stablelm-tuned-alpha-7b`, etc.)
53
+- InternLM (`internlm/internlm-7b`, `internlm/internlm-chat-7b`, etc.)
54
- LLaMA & LLaMA-2 (`meta-llama/Llama-2-70b-hf`, `lmsys/vicuna-13b-v1.3`, `young-geng/koala`, `openlm-research/open_llama_13b`, etc.)
55
- MPT (`mosaicml/mpt-7b`, `mosaicml/mpt-30b`, etc.)
56
- OPT (`facebook/opt-66b`, `facebook/opt-iml-max-30b`, etc.)
57
+- Qwen (`Qwen/Qwen-7B`, `Qwen/Qwen-7B-Chat`, etc.)
58
59
Install vLLM with pip or [from source](https://vllm.readthedocs.io/en/latest/getting_started/installation.html#build-from-source):
60
docs/source/models/supported_models.rst
+9
@@ -14,6 +14,9 @@ Alongside each architecture, we include some popular models that use it.
14
* - Architecture
15
- Models
16
- Example HuggingFace Models
17
+ * - :code:`AquilaForCausalLM`
18
+ - Aqualia
19
+ - :code:`BAAI/Aquila-7B`, :code:`BAAI/AquilaChat-7B`, etc.
20
* - :code:`BaiChuanForCausalLM`
21
- Baichuan
22
- :code:`baichuan-inc/Baichuan-7B`, :code:`baichuan-inc/Baichuan-13B-Chat`, etc.
@@ -35,6 +38,9 @@ Alongside each architecture, we include some popular models that use it.
38
* - :code:`GPTNeoXForCausalLM`
39
- GPT-NeoX, Pythia, OpenAssistant, Dolly V2, StableLM
40
- :code:`EleutherAI/gpt-neox-20b`, :code:`EleutherAI/pythia-12b`, :code:`OpenAssistant/oasst-sft-4-pythia-12b-epoch-3.5`, :code:`databricks/dolly-v2-12b`, :code:`stabilityai/stablelm-tuned-alpha-7b`, etc.
41
+ * - :code:`InternLMForCausalLM`
42
+ - InternLM
43
+ - :code:`internlm/internlm-7b`, :code:`internlm/internlm-chat-7b`, etc.
44
* - :code:`LlamaForCausalLM`
45
- LLaMA, LLaMA-2, Vicuna, Alpaca, Koala, Guanaco
46
- :code:`meta-llama/Llama-2-13b-hf`, :code:`openlm-research/open_llama_13b`, :code:`lmsys/vicuna-13b-v1.3`, :code:`young-geng/koala`, :code:`JosephusCheung/Guanaco`, etc.
@@ -44,6 +50,9 @@ Alongside each architecture, we include some popular models that use it.
50
* - :code:`OPTForCausalLM`
51
- OPT, OPT-IML
52
- :code:`facebook/opt-66b`, :code:`facebook/opt-iml-max-30b`, etc.
53
+ * - :code:`OPTForCausalLM`
54
+ - Qwen
55
+ - :code:`Qwen/Qwen-7B`, :code:`Qwen/Qwen-7B-Chat`, etc.
56
57
If your model uses one of the above model architectures, you can seamlessly run your model with vLLM.
58
Otherwise, please refer to :ref:`Adding a New Model <adding_a_new_model>` for instructions on how to implement support for your model.