Terminal Bench 4 Released: Removes 8 Tasks for Saturation and Refusals
cwolferesearch · x · 2026-08-30
Terminal Bench 4 has been released just a month after version 3, marking a shift to a continuous-style benchmark. No new tasks were added; instead, 8 tasks were removed based on feedback from TB3 to reduce noise. Removal reasons include task saturation (100% solve rate by latest models), high refusal rates, existing public solutions, and quality issues. The update focuses on fixing known issues and improving evaluation infrastructure.
Related event: Terminal-Bench 4.0 Released: Opus 5 Leads, GLM 5.3 Top Open-Source Model(14 posts)→
More from Research
- WorldMemArena Accepted at NeurIPS: Memory Storage ≠ Memory Use — xwang_lk · 2026-10-16
- Mech Interp Veteran Rethinks Alignment, Misgeneralization, CoT Oversight — akbirthko · 2026-10-16
- WEDGE paper generates training environments conditioned on navigation agent failures — _akhaliq · 2026-10-16
- RExBench: best coding agent completes only 33% of autonomous research extensions — najoungkim · 2026-10-16
- Google's CLASHarness benchmarks multi-agent conflicts in computer use, finds frontier VLMs struggle — google · 2026-10-16
- Tsinghua team proves vanilla flow-matching sampling can't exceed constant-depth expressive power — SQZ · 2026-10-16