Ornith-1.5 开源发布，397B 多项基准超 Claude Opus
Another brilliant open-source model release.
Another brilliant open-source model release.
又一个出色的开源模型发布。
Ornith-1.5, a family of open-source LLMs spanning 9B Dense, 35B MoE, and 397B MoE,
Ornith-1.5，一个开源LLM系列，涵盖9B密集、35B MoE和397B MoE。
Only 397B weights, but 4 wins over Claude Opus 4.8:
仅397B权重，但在4项上胜过Claude Opus 4.8：
Terminal-Bench 2.1 (86.1 v 85), SWE-bench Verified (86 v 85.8), WideSearch (80.8 v 72.9), BrowseComp (86.6 v 84.3).
Terminal-Bench 2.1（86.1对85）、SWE-bench Verified（86对85.8）、WideSearch（80.8对72.9）、BrowseComp（86.6对84.3）。
The architectural jump for Ornith-1.5 is that it effectively puts task generation, agent design, and solution generation inside the same reinforcement-learning loop. Improvement in one part can feed the other 2.
Ornith-1.5的架构飞跃在于，它有效地将任务生成、代理设计和解决方案生成置于同一个强化学习循环中。一个部分的改进可以反馈给其他两个部分。
i.e. the model proposes its own tasks, writes the scaffold that will grade them, produces solution rollouts, and propagates reward back through all 3 stages with GRPO.
即，模型提出自己的任务，编写用于评分的脚手架，生成解决方案的展开，并通过GRPO将奖励传播回所有三个阶段。
So the self-generated curriculum removes many human task design. Ornith begins with an environment or codebase plus high-level task instructions, then searches near the model’s current capability frontier.
因此，自我生成的课程减少了许多人工任务设计。 Ornith从一个环境或代码库加上高层任务指令开始，然后在模型当前能力前沿附近进行搜索。
That makes repositories and other verifiable environments more valuable training inputs than static examples alone. Labs with richer places to generate and verify new tasks could improve faster even without starting from the largest human-built dataset.
这使得代码库和其他可验证环境成为比静态示例更有价值的训练输入。拥有更丰富环境来生成和验证新任务的实验室，即使不从最大的人工构建数据集开始，也能更快地改进。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力