Grok 4.6 tested on bug bench: outperforms predecessor, becomes new default
PawelHuryn · x · 2026-08-13
An hour after the release of Grok 4.6, a developer tested it on a benchmark of 105 hidden bugs in real repos. Grok 4.6 fixed 27 bugs (including 15 non-planted), outperforming Grok 4.5 (17 bugs) but slightly trailing Fable 5 (29 bugs). The author noted it as the best combination of time, value, and cost, making it their new default model.
Related event: xAI Releases Grok 4.6 with Top Performance and Same Pricing(92 posts)→
More from coding & agent
- DIRECT: retrieval-based next-question suggestions for conversational agents — burkov · 2026-10-16
- Claude Code and Codex now autocomplete the user's next action — tokenbender · 2026-10-16
- Hermes Agent ships official Reverse Engineer Anything plugin with one-command setup — Teknium · 2026-10-16
- MLExpert: 35B MoE Agent Trained to Autonomously Handle Machine Learning Engineering — Shangheng Du · 2026-10-16
- Cowork-Gym: Alibaba Synthesizes 64.9K Agent Work Environments, Boosting Qwen by 8.7 Points — Anni Zou · 2026-10-16
- MMSearch-OPD: Joint RL and On-Policy Distillation for Long-Horizon Multimodal Search Agents — PekingUniversity · 2026-10-16