From 66d4edda411a868d242124eb179c6148d6d73dde Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E6=9F=B4=E7=B1=B3=E6=B2=B9=E7=9B=90=E7=9A=84=E6=A2=A6?= =?UTF-8?q?=E6=83=B3?= <107172917+chai-mi@users.noreply.github.com> Date: Fri, 24 Jan 2025 00:53:32 +0800 Subject: [PATCH] =?UTF-8?q?=F0=9F=92=84=20style:=20Add=20HuggingFace=20Mod?= =?UTF-8?q?el:=20DeepSeek=20R1=20(#5564)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- src/config/aiModels/huggingface.ts | 9 ++++++++- src/config/modelProviders/huggingface.ts | 8 +++++++- 2 files changed, 15 insertions(+), 2 deletions(-) diff --git a/src/config/aiModels/huggingface.ts b/src/config/aiModels/huggingface.ts index 02d360b756fb..602d2bb355ca 100644 --- a/src/config/aiModels/huggingface.ts +++ b/src/config/aiModels/huggingface.ts @@ -5,7 +5,6 @@ const huggingfaceChatModels: AIChatModelCard[] = [ contextWindowTokens: 32_768, description: 'Mistral AI的指令调优模型', displayName: 'Mistral 7B Instruct v0.3', - enabled: true, id: 'mistralai/Mistral-7B-Instruct-v0.3', type: 'chat', }, @@ -34,6 +33,7 @@ const huggingfaceChatModels: AIChatModelCard[] = [ contextWindowTokens: 32_768, description: 'Qwen QwQ 是由 Qwen 团队开发的实验研究模型,专注于提升AI推理能力。', displayName: 'QwQ 32B Preview', + enabled: true, id: 'Qwen/QwQ-32B-Preview', type: 'chat', }, @@ -49,6 +49,13 @@ const huggingfaceChatModels: AIChatModelCard[] = [ id: 'NousResearch/Hermes-3-Llama-3.1-8B', type: 'chat', }, + { + contextWindowTokens: 16_384, + displayName: 'DeepSeek R1', + enabled: true, + id: 'deepseek-ai/DeepSeek-R1-Distill-Qwen-32B', + type: 'chat', + }, ]; export const allModels = [...huggingfaceChatModels]; diff --git a/src/config/modelProviders/huggingface.ts b/src/config/modelProviders/huggingface.ts index ace8370720c0..a172dc515f88 100644 --- a/src/config/modelProviders/huggingface.ts +++ b/src/config/modelProviders/huggingface.ts @@ -6,7 +6,6 @@ const HuggingFace: ModelProviderCard = { contextWindowTokens: 32_768, description: 'Mistral AI的指令调优模型', displayName: 'Mistral 7B Instruct v0.3', - enabled: true, id: 'mistralai/Mistral-7B-Instruct-v0.3', }, { @@ -31,6 +30,7 @@ const HuggingFace: ModelProviderCard = { contextWindowTokens: 32_768, description: 'Qwen QwQ 是由 Qwen 团队开发的实验研究模型,专注于提升AI推理能力。', displayName: 'QwQ 32B Preview', + enabled: true, id: 'Qwen/QwQ-32B-Preview', }, { @@ -43,6 +43,12 @@ const HuggingFace: ModelProviderCard = { displayName: 'Hermes 3 Llama 3.1 8B', id: 'NousResearch/Hermes-3-Llama-3.1-8B', }, + { + contextWindowTokens: 16_384, + displayName: 'DeepSeek R1', + enabled: true, + id: 'deepseek-ai/DeepSeek-R1-Distill-Qwen-32B', + }, ], checkModel: 'mistralai/Mistral-7B-Instruct-v0.2', description: