PulseAugur
中
实时 20:35:20
English(EN) DeepReinforce Releases Ornith-1.0: An Open-Source Coding Model Family That Learns Its Own RL Scaffolds

DeepReinforce 发布 Ornith-1.0 开源编码模型，可学习 RL 脚手架

作者 PulseAugur 编辑部 · [2 个来源] ·

DeepReinforce 推出了 Ornith-1.0，这是一个在 MIT 许可下提供的开源编码模型家族。这些模型基于 Gemma 4 和 Qwen 3.5 构建，专为代理编码任务设计，并在训练过程中独特地学习自身的强化学习脚手架。最大的模型 Ornith-1.0-397B 在 SWE-Bench Verified 基准测试中取得了 82.4% 的优异成绩。 AI

影响 此次发布为训练编码代理提供了一种新颖的方法，有可能提高它们在没有固定约束的情况下学习和适应的能力。

排序理由 具有新颖的自脚手架 RL 功能的新模型家族的开源发布。[lever_c_demoted from frontier_release: ic=2 ai=1.0]

在 MarkTechPost 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

DeepReinforce 发布 Ornith-1.0 开源编码模型，可学习 RL 脚手架

报道来源 [2]

  1. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    DeepReinforce Releases Ornith-1.0: An Open-Source Coding Model Family That Learns Its Own RL Scaffolds

    <p>DeepReinforce released Ornith-1.0, an open-source coding model family built on Gemma 4 and Qwen 3.5. Instead of a fixed harness, the model learns its own scaffold during reinforcement learning. The 397B flagship reports 82.4 on SWE-Bench Verified, with all weights under the MI…

  2. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    DeepReinforce has released Ornith-1.0, an open-source coding model family built on Gemma 4 and Qwen 3.5 that learns its own RL scaffolds during training. The 39

    DeepReinforce has released Ornith-1.0, an open-source coding model family built on Gemma 4 and Qwen 3.5 that learns its own RL scaffolds during training. The 397B model achieves 82.4% on SWE-Bench Verified and is available under the MIT license. https://www. marktechpost.com/2026…

相关实体

相关话题