English(EN) IBM has released Granite 4.2, a family of open reasoning language models with native chain-of-thought and an agentic RL block that trains models to edit code, d

IBM 发布具有代理功能的 Granite 4.2 开放推理模型

作者 PulseAugur 编辑部 · [1 个来源] · 2026-08-26 05:51

IBM 推出了 Granite 4.2，一个新系列的开放推理语言模型。这些模型具有内置的链式思考能力和一个代理式强化学习模块，旨在训练模型在安全环境中执行代码编辑、终端操作和网络搜索等任务。这个拥有300亿参数的模型在 SWE-Bench Verified 基准测试中获得了 57.00 分。 AI

影响通过内置的推理和代理功能，增强了企业在代码编辑和网络任务方面的 AI 能力。

排序理由前沿实验室模型发布，带有系统卡 [lever_c_demoted from frontier_release: ic=1 ai=1.0]

在 Mastodon — fosstodon.org 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 19 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Significant; 前沿实验室模型发布，带有系统卡 [lever_c_demoted from frontier_release: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: model release, product; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-08-26 05:51

IBM 发布 Granite 4.2，一个包含原生链式思考和训练模型编辑代码的代理式 RL 模块的开放推理语言模型家族

IBM has released Granite 4.2, a family of open reasoning language models with native chain-of-thought and an agentic RL block that trains models to edit code, drive terminals and run web searches in sandboxed environments. The 30B model scores 57.00 on SWE-Bench Verified. https:/…

本文如何被排名

报道来源 [1]

IBM 发布 Granite 4.2，一个包含原生链式思考和训练模型编辑代码的代理式 RL 模块的开放推理语言模型家族

相关实体

相关话题