#coding benchmarks

16 stories · 0 in 7 days · 2 in 30 days
Stories, products and papers tagged "coding benchmarks", newest first.
Oct 21
PapersBuildersMulti-sourceSelected10:17
Microsoft Paper Co-Evolves RL Harness and Qwen3.5 Model

Co-evolving the harness and the model appears to unlock gains that optimizing either alone cannot reach, suggesting the harness itself is a first-class lever in RL training. Teams training coding agents can copy the loop to squeeze more performance from mid-size open models like Qwen3.5-35B-A3B. The distillation trick of using Opus 4.8 as the harness proposer also offers a cheaper route to borrow frontier-model judgment.

Event
DAIR.AI@dair_aiSource
Oct 18
Sep 9
Sep 6
Sep 3
Jul 21
Jul 19
Jul 9
Jun 28
Jun 26
Jun 17
Jun 12
ProductsBuildersMulti-source13:31
Chinese contentMythos 上线:FrontierCode 成为新编程基准,Opus 4.8 和 GPT 5.5 在 FC Diamond 上不随努力扩展

Mythos 的 FrontierCode 基准揭示了当前顶级模型在长任务上的扩展瓶颈,做 AI 编程评估或开发长流程自动化的团队值得关注,可以直接在 Devin 中体验。

Event
swyx (AI Engineer)@swyxSource
Jun 9
May 30