GPT-6.1 Terra Benchmarks: Sweeps All Six Tests, SWE-bench at 76.2%
mark_k · x · 2026-10-27
Benchmark results for OpenAI's GPT-6.1 Terra are out, topping all six evaluated benchmarks against GPT-5.6 Terra, GPT-6 Astra, GPT-6.1 Sol, Claude Opus 5.5, and Gemini 2.5 Pro:
- Terminal-Bench 4.0 (coding): 50.3%, well ahead of GPT-6.1 Sol's 34.1%
- Factuality (internal eval, lower error is better): 16.7% vs GPT-5.6 Terra's 53.6%
- Agents' Last Exam (computer use): 41.6%
- SWE-bench Verified: 76.2% vs Claude Opus 5.5's 71.6%
- BrowseComp (web browsing): 62.3%
- Synthetic Difficult Questions: 74.8%
Note: results come from OpenAI's internal evaluations and selected public benchmarks; test setups may differ across benchmarks and models.
More from Models
- Poster surprised that v4.1 flash gets more accurate when asked to control its CoT — dfrsrchtwts · 2026-10-27
- Dev claims Anthropic is baking 'sabotage' into model behavior for 'wrong' engineering — cephaloform · 2026-10-27
- Gemini fails at listing unread emails, tells user to search Gmail manually — gethackteam · 2026-10-27
- Meta's Muse Spark Open goes live on Applied Compute for async RL post-training — ypatil125 · 2026-10-27
- Moonshot curbs hacking ability in Kimi K3.1 Pro, its strongest 3T-class open-weight model — pstAsiatech · 2026-10-27
- Would you want an argumentative LLM that never backs down, even under prompt injection? — GodComplecs · 2026-10-27