Multiverse 发布欧洲大模型 Quasar 438B
Quasar sets a new benchmark for European AI, outperforming comparable European models on seven of eight selected Artificial Analysis evaluations. ...
Quasar 为欧洲 AI 树立新基准，在八项选定的 Artificial Analysis 评测中有七项优于可比的欧洲模型。……
原文来源：Multiverse（multiversecomputing.com）。本页摘要与要点整理自公开来源，版权归原作者所有，完整内容请阅读原文。
Check if files were made with ClaudeAnthropic 上线 Claude 内容检测工具
Anthropic · 2026-09-02
Introducing Gemini 3.8 Flash and 3.8 Flash CyberGoogle 发布 Gemini 3.8 Flash 与 3.8 Flash Cyber
Google DeepMind · 2026-09-03
BenchMIRT: What are LLM benchmarks actually measuring?Ai2 推出 BenchMIRT：在单题层面审计大模型基准
Hugging Face · 2026-09-02
Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO StepsHugging Face 教程：100 步 GRPO 提升 350M 小模型结构化输出能力
Hugging Face · 2026-09-03
Copilot code review can now approve pull requestsCopilot 代码审查现可批准 Pull Request
GitHub Changelog · 2026-09-02
Introducing Claude Fable 5.1 and Claude Mythos 5.1Anthropic 发布 Claude Fable 5.1 与 Claude Mythos 5.1
Anthropic · 2026-09-02