Terminal-Bench 4.0 Updates Tasks and Fixes; Opus 5 and GLM 5.3 Lead
heypearlai · x · 2026-08-29
Terminal-Bench 4.0 is released, featuring recalibrated time and compute resources (flat 8-hour timeout), 19 fixed tasks to address flakiness, and the removal of 8 saturated or problematic tasks. This version significantly reduces agent timeouts and noise. Opus 5 and GLM 5.3 currently lead the leaderboard, with the open-source GLM 5.3 keeping pace with Opus 5 on efficiency. Notably, Sonnet 5 showed high token consumption (21.6B) and variance in execution time.
Related event: Terminal-Bench 4.0 Released: Opus 5 Leads, GLM 5.3 Top Open-Source Model(14 posts)→
More from coding & agent
- Vercel Goes Full Backend: Python GA, Containers, and Daemons — cramforce · 2026-10-16
- WorldMemArena Accepted at NeurIPS: Memory Storage ≠ Memory Use — xwang_lk · 2026-10-16
- Vercel Ship Demos Container Deploys for Go, Ruby, Java and More — edwin · 2026-10-16
- TRL adds day-0 support for fine-tuning Muse Spark Open's 816B MoE with LoRA on 64 H100s — QGallouedec · 2026-10-16
- Databricks engineers land impactful PRs on day 2 with AI dev tools, vs week 3 before — Yuchenj_UW · 2026-10-16
- Perplexity open-sources SPACE, the sandbox platform powering its Computer agents — AravSrinivas · 2026-10-16