From 34ba81716add881eed4c33c8acd6b5c64380d980 Mon Sep 17 00:00:00 2001 From: Uranus <109661872+UranusSeven@users.noreply.github.com> Date: Thu, 17 Aug 2023 20:58:15 +0800 Subject: [PATCH] DOC: update builtin models (#365) --- README.md | 46 ++++++++++++++++++++++++++-------------------- README_ja_JP.md | 47 ++++++++++++++++++++++++++--------------------- README_zh_CN.md | 46 ++++++++++++++++++++++++++-------------------- 3 files changed, 78 insertions(+), 61 deletions(-) diff --git a/README.md b/README.md index afb06da2dc..5662b9a4d1 100644 --- a/README.md +++ b/README.md @@ -193,26 +193,32 @@ To view the builtin models, run the following command: $ xinference list --all ``` -| Name | Language | Ability | -|------------------|--------------|-----------------------| -| baichuan | ['en', 'zh'] | ['embed', 'generate'] | -| baichuan-chat | ['en', 'zh'] | ['embed', 'chat'] | -| chatglm | ['en', 'zh'] | ['embed', 'chat'] | -| chatglm2 | ['en', 'zh'] | ['embed', 'chat'] | -| chatglm2-32k | ['en', 'zh'] | ['embed', 'chat'] | -| falcon | ['en'] | ['embed', 'generate'] | -| falcon-instruct | ['en'] | ['embed', 'chat'] | -| gpt-2 | ['en'] | ['generate'] | -| llama-2 | ['en'] | ['embed', 'generate'] | -| llama-2-chat | ['en'] | ['embed', 'chat'] | -| opt | ['en'] | ['embed', 'generate'] | -| orca | ['en'] | ['embed', 'chat'] | -| qwen-chat | ['en', 'zh'] | ['embed', 'chat'] | -| starcoder | ['en'] | ['generate'] | -| starcoderplus | ['en'] | ['embed', 'generate'] | -| starchat-beta | ['en'] | ['embed', 'chat'] | -| vicuna-v1.3 | ['en'] | ['embed', 'chat'] | -| wizardlm-v1.0 | ['en'] | ['embed', 'chat'] | +| Name | Language | Ability | +|------------------|---------------|------------------------| +| baichuan | ['en', 'zh'] | ['embed', 'generate'] | +| baichuan-chat | ['en', 'zh'] | ['embed', 'chat'] | +| chatglm | ['en', 'zh'] | ['embed', 'chat'] | +| chatglm2 | ['en', 'zh'] | ['embed', 'chat'] | +| chatglm2-32k | ['en', 'zh'] | ['embed', 'chat'] | +| falcon | ['en'] | ['embed', 'generate'] | +| falcon-instruct | ['en'] | ['embed', 'chat'] | +| gpt-2 | ['en'] | ['generate'] | +| internlm | ['en', 'zh'] | ['embed', 'generate'] | +| internlm-chat | ['en', 'zh'] | ['embed', 'chat'] | +| internlm-chat-8k | ['en', 'zh'] | ['embed', 'chat'] | +| llama-2 | ['en'] | ['embed', 'generate'] | +| llama-2-chat | ['en'] | ['embed', 'chat'] | +| opt | ['en'] | ['embed', 'generate'] | +| orca | ['en'] | ['embed', 'chat'] | +| qwen-chat | ['en', 'zh'] | ['embed', 'chat'] | +| starchat-beta | ['en'] | ['embed', 'chat'] | +| starcoder | ['en'] | ['generate'] | +| starcoderplus | ['en'] | ['embed', 'generate'] | +| vicuna-v1.3 | ['en'] | ['embed', 'chat'] | +| vicuna-v1.5 | ['en'] | ['embed', 'chat'] | +| vicuna-v1.5-16k | ['en'] | ['embed', 'chat'] | +| wizardlm-v1.0 | ['en'] | ['embed', 'chat'] | +| wizardmath-v1.0 | ['en'] | ['embed', 'chat'] | For in-depth details on the built-in models, please refer to [built-in models](https://inference.readthedocs.io/en/latest/models/builtin/index.html). diff --git a/README_ja_JP.md b/README_ja_JP.md index 16a412daeb..ae135f574e 100644 --- a/README_ja_JP.md +++ b/README_ja_JP.md @@ -171,28 +171,33 @@ model.chat( $ xinference list --all ``` -### ggmlv3 モデル - -| Name | Type | Language | Format | Size (in billions) | Quantization | -|---------------|------------------|----------|---------|--------------------|-----------------------------------------| -| llama-2 | Foundation Model | en | ggmlv3 | 7, 13 | 'q2_K', 'q3_K_L', ... , 'q6_K', 'q8_0' | -| baichuan | Foundation Model | en, zh | ggmlv3 | 7 | 'q2_K', 'q3_K_L', ... , 'q6_K', 'q8_0' | -| llama-2-chat | RLHF Model | en | ggmlv3 | 7, 13, 70 | 'q2_K', 'q3_K_L', ... , 'q6_K', 'q8_0' | -| chatglm | SFT Model | en, zh | ggmlv3 | 6 | 'q4_0', 'q4_1', 'q5_0', 'q5_1', 'q8_0' | -| chatglm2 | SFT Model | en, zh | ggmlv3 | 6 | 'q4_0', 'q4_1', 'q5_0', 'q5_1', 'q8_0' | -| wizardlm-v1.0 | SFT Model | en | ggmlv3 | 7, 13, 33 | 'q2_K', 'q3_K_L', ... , 'q6_K', 'q8_0' | -| wizardlm-v1.1 | SFT Model | en | ggmlv3 | 13 | 'q2_K', 'q3_K_L', ... , 'q6_K', 'q8_0' | -| vicuna-v1.3 | SFT Model | en | ggmlv3 | 7, 13 | 'q2_K', 'q3_K_L', ... , 'q6_K', 'q8_0' | -| orca | SFT Model | en | ggmlv3 | 3, 7, 13 | 'q4_0', 'q4_1', 'q5_0', 'q5_1', 'q8_0' | - -### pytorch モデル - -| Name | Type | Language | Format | Size (in billions) | Quantization | -|---------------|------------------|----------|---------|--------------------|--------------------------| -| baichuan | Foundation Model | en, zh | pytorch | 7, 13 | '4-bit', '8-bit', 'none' | -| baichuan-chat | SFT Model | en, zh | pytorch | 13 | '4-bit', '8-bit', 'none' | -| vicuna-v1.3 | SFT Model | en | pytorch | 7, 13, 33 | '4-bit', '8-bit', 'none' | +| Name | Language | Ability | +|------------------|---------------|------------------------| +| baichuan | ['en', 'zh'] | ['embed', 'generate'] | +| baichuan-chat | ['en', 'zh'] | ['embed', 'chat'] | +| chatglm | ['en', 'zh'] | ['embed', 'chat'] | +| chatglm2 | ['en', 'zh'] | ['embed', 'chat'] | +| chatglm2-32k | ['en', 'zh'] | ['embed', 'chat'] | +| falcon | ['en'] | ['embed', 'generate'] | +| falcon-instruct | ['en'] | ['embed', 'chat'] | +| gpt-2 | ['en'] | ['generate'] | +| internlm | ['en', 'zh'] | ['embed', 'generate'] | +| internlm-chat | ['en', 'zh'] | ['embed', 'chat'] | +| internlm-chat-8k | ['en', 'zh'] | ['embed', 'chat'] | +| llama-2 | ['en'] | ['embed', 'generate'] | +| llama-2-chat | ['en'] | ['embed', 'chat'] | +| opt | ['en'] | ['embed', 'generate'] | +| orca | ['en'] | ['embed', 'chat'] | +| qwen-chat | ['en', 'zh'] | ['embed', 'chat'] | +| starchat-beta | ['en'] | ['embed', 'chat'] | +| starcoder | ['en'] | ['generate'] | +| starcoderplus | ['en'] | ['embed', 'generate'] | +| vicuna-v1.3 | ['en'] | ['embed', 'chat'] | +| vicuna-v1.5 | ['en'] | ['embed', 'chat'] | +| vicuna-v1.5-16k | ['en'] | ['embed', 'chat'] | +| wizardlm-v1.0 | ['en'] | ['embed', 'chat'] | +| wizardmath-v1.0 | ['en'] | ['embed', 'chat'] | **注**: - Xinference は自動的にモデルをダウンロードし、デフォルトでは `${USER}/.xinference/cache` の下に保存されます。 diff --git a/README_zh_CN.md b/README_zh_CN.md index 3eae2df314..c209e3279c 100644 --- a/README_zh_CN.md +++ b/README_zh_CN.md @@ -174,26 +174,32 @@ model.chat( $ xinference list --all ``` -| Name | Language | Ability | -|------------------|--------------|-----------------------| -| baichuan | ['en', 'zh'] | ['embed', 'generate'] | -| baichuan-chat | ['en', 'zh'] | ['embed', 'chat'] | -| chatglm | ['en', 'zh'] | ['embed', 'chat'] | -| chatglm2 | ['en', 'zh'] | ['embed', 'chat'] | -| chatglm2-32k | ['en', 'zh'] | ['embed', 'chat'] | -| falcon | ['en'] | ['embed', 'generate'] | -| falcon-instruct | ['en'] | ['embed', 'chat'] | -| gpt-2 | ['en'] | ['generate'] | -| llama-2 | ['en'] | ['embed', 'generate'] | -| llama-2-chat | ['en'] | ['embed', 'chat'] | -| opt | ['en'] | ['embed', 'generate'] | -| orca | ['en'] | ['embed', 'chat'] | -| qwen-chat | ['en', 'zh'] | ['embed', 'chat'] | -| starcoder | ['en'] | ['generate'] | -| starcoderplus | ['en'] | ['embed', 'generate'] | -| starchat-beta | ['en'] | ['embed', 'chat'] | -| vicuna-v1.3 | ['en'] | ['embed', 'chat'] | -| wizardlm-v1.0 | ['en'] | ['embed', 'chat'] | +| Name | Language | Ability | +|------------------|---------------|------------------------| +| baichuan | ['en', 'zh'] | ['embed', 'generate'] | +| baichuan-chat | ['en', 'zh'] | ['embed', 'chat'] | +| chatglm | ['en', 'zh'] | ['embed', 'chat'] | +| chatglm2 | ['en', 'zh'] | ['embed', 'chat'] | +| chatglm2-32k | ['en', 'zh'] | ['embed', 'chat'] | +| falcon | ['en'] | ['embed', 'generate'] | +| falcon-instruct | ['en'] | ['embed', 'chat'] | +| gpt-2 | ['en'] | ['generate'] | +| internlm | ['en', 'zh'] | ['embed', 'generate'] | +| internlm-chat | ['en', 'zh'] | ['embed', 'chat'] | +| internlm-chat-8k | ['en', 'zh'] | ['embed', 'chat'] | +| llama-2 | ['en'] | ['embed', 'generate'] | +| llama-2-chat | ['en'] | ['embed', 'chat'] | +| opt | ['en'] | ['embed', 'generate'] | +| orca | ['en'] | ['embed', 'chat'] | +| qwen-chat | ['en', 'zh'] | ['embed', 'chat'] | +| starchat-beta | ['en'] | ['embed', 'chat'] | +| starcoder | ['en'] | ['generate'] | +| starcoderplus | ['en'] | ['embed', 'generate'] | +| vicuna-v1.3 | ['en'] | ['embed', 'chat'] | +| vicuna-v1.5 | ['en'] | ['embed', 'chat'] | +| vicuna-v1.5-16k | ['en'] | ['embed', 'chat'] | +| wizardlm-v1.0 | ['en'] | ['embed', 'chat'] | +| wizardmath-v1.0 | ['en'] | ['embed', 'chat'] | 更多信息请参考 [内置模型](https://inference.readthedocs.io/en/latest/models/builtin/index.html)。