追踪开放权重模型、开源 AI 工具和可复核的仓库更新。
# vLLM v0.23.0 Release Notes Please note that Minimax M3 is not yet supported in this version. Please follow [vLLM recipe](https://recipes.vllm.ai/MiniMaxAI/MiniMax-M3) for usage guides for M3. ## Highlights This release features 408 commits from 200 contribut
来源已更新，查看详情及原始发布记录。
## Highlights This release features 8 commits from 6 contributors (1 new)! v0.22.1 is a patch release on top of v0.22.0 with targeted bug fixes plus a couple of additions: new model support for JetBrains' Mellum v2, zentorch-accelerated quantized linear infere
### Summary of major features and improvements * #### More GenAI coverage and framework integrations to minimize code changes * New models supported: Gemma 4 E2B and Gemma 4 E4B * Only on CPUs & GPUs: Qwen3-Coder-Next, Qwen3.5, Qwen3.6, Trinity-mini, LFM2-24B-
模型仓库动态。仓库创建：2026-05-21T04:15:56.000Z。最后修改：2026-05-28T08:07:27.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。 任务：image-text-to-text。
v0.5.12.post1 is a stability patch on top of v0.5.12. It cherry-picks 12 fixes — primarily for DeepSeek V4 — onto the release branch. # Bug Fixes ## DeepSeek V4 * DSV4-Pro emits garbled text during single-token decode on B200/B300 (fix `deep_gemm` UE8M0 scale-
模型仓库动态。仓库创建：2026-04-27T03:39:37.000Z。最后修改：2026-05-13T12:19:26.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。
模型仓库动态。仓库创建：2026-04-27T03:39:38.000Z。最后修改：2026-05-13T12:19:23.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。
模型仓库动态。仓库创建：2026-04-27T03:39:40.000Z。最后修改：2026-05-13T12:19:20.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。
模型仓库动态。仓库创建：2026-04-03T09:28:47.000Z。最后修改：2026-05-13T08:10:58.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。 任务：text-generation。
来源标注日期：2026-04-24（未提供具体时刻）。 The DeepSeek API now supports V4-Pro and V4-Flash, available via both the OpenAI ChatCompletions interface and the Anthropic interface. To access the new models, the base_url remains unchanged, and the model parameter should be set
模型仓库动态。仓库创建：2026-04-03T09:29:04.000Z。最后修改：2026-04-16T06:12:42.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。 任务：text-generation。
### Summary of major features and improvements * #### More GenAI coverage and framework integrations to minimize code changes * New models supported on CPUs & GPUs: Qwen3 VL * New models supported on CPUs: GPT-OSS 120B * Preview: Introducing the OpenVINO backe
模型仓库动态。仓库创建：2025-12-09T09:07:41.000Z。最后修改：2026-04-07T12:55:32.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。 任务：automatic-speech-recognition。
模型仓库动态。仓库创建：2026-02-11T04:05:41.000Z。最后修改：2026-04-05T07:51:12.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。 任务：text-generation。
Gemma 4: Our most intelligent open models to date, purpose-built for advanced reasoning and agentic workflows.
### Summary of major features and improvements * #### More GenAI coverage and framework integrations to minimize code changes * New models supported on CPUs & GPUs: GPT-OSS-20B, MiniCPM-V-4_5-8B, and MiniCPM-o-2.6 * New models supported on NPUs: MiniCPM-o-2.6
* Preview: NPU compiler integration with the NPU plugin enables ahead-of-time and on-device compilation without relying on OEM driver updates. This feature is enabled by default in this release package. * Known Issues * Component: optimum; ID: 179936 Descripti
模型仓库动态。仓库创建：2026-01-19T06:28:10.000Z。最后修改：2026-01-29T08:06:19.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。 任务：text-generation。
### Summary of major features and improvements * #### More GenAI coverage and framework integrations to minimize code changes * **New models supported:** * On CPUs & GPUs: **Qwen3-Embedding-0.6B, Qwen3-Reranker-0.6B, Mistral-Small-24B-Instruct-2501.** * On NP
Open interpretability tools for language models are now available across the entire Gemma 3 family with the release of Gemma Scope 2.
来源标注日期：2025-12-01（未提供具体时刻）。 Both deepseek-chat and deepseek-reasoner have been upgraded to DeepSeek-V3.2. deepseek-chat corresponds to DeepSeek-V3.2's non-thinking mode deepseek-reasoner corresponds to DeepSeek-V3.2's thinking mode DeepSeek-V3.2-Speciale is se
模型仓库动态。仓库创建：2024-07-21T19:18:13.000Z。最后修改：2025-11-12T21:27:00.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。 任务：text-classification。
This release is meant to fix the following issues (regressions / silent correctness): ### Tracked Regressions Significant Memory Regression in F.conv3d with bfloat16 Inputs in PyTorch 2.9.0 ([#166643](https://github.com/pytorch/pytorch/issues/166643)) This rel
模型仓库动态。仓库创建：2025-10-08T07:23:59.000Z。最后修改：2025-10-24T02:07:00.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。 任务：text-generation。
来源标注日期：2025-09-29（未提供具体时刻）。 Both deepseek-chat and deepseek-reasoner have been upgraded to DeepSeek-V3.2-Exp. deepseek-chat corresponds to DeepSeek-V3.2-Exp's non-thinking mode deepseek-reasoner corresponds to DeepSeek-V3.2-Exp's thinking mode For more details
来源标注日期：2025-09-22（未提供具体时刻）。 Both deepseek-chat and deepseek-reasoner have been upgraded to DeepSeek-V3.1-Terminus. deepseek-chat corresponds to DeepSeek-V3.1-Terminus's non-thinking mode, while deepseek-reasoner corresponds to its thinking mode. This update ma
来源标注日期：2025-08-21（未提供具体时刻）。 Both deepseek-chat and deepseek-reasoner have been upgraded to DeepSeek-V3.1. deepseek-chat corresponds to DeepSeek-V3.1's non-thinking mode, while deepseek-reasoner corresponds to its thinking mode. Key updates in DeepSeek-V3.1: Hy
模型仓库动态。仓库创建：2024-04-17T09:35:12.000Z。最后修改：2025-06-18T23:49:51.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。 任务：text-generation。
模型仓库动态。仓库创建：2024-04-17T09:34:54.000Z。最后修改：2025-06-18T23:49:26.000Z。仓库修改不等同于模型正式发布。 数据经第三方 Hugging Face 镜像采集，需以原始模型页面复核。 任务：text-generation。
