GPT-6.1 Terra Slashes Factual Error Rate from 22.3% to 4.3%
Benchmark data shows GPT-6.1 Terra cuts its factual error rate on hard prompts from 22.3% to 4.3%, a near-81% reduction, while accuracy also improves by 2 points.
2026-10-27 ~ 2026-10-27 · 3 related posts
- Episode 1: Microsoft PR leak reveals unreleased OpenAI model GPT-6.1 Terra(2026-10-19, 6 posts)
- Episode 2: Suspected OpenAI model GPT-6.1 Terra surfaces across multiple channels ahead of likely launch(2026-10-25, 10 posts)
- Episode 3: Report: OpenAI pauses frontier RL training, shifts to smaller cheaper models(2026-10-25, 2 posts)
- Episode 4: OpenAI releases GPT-6.1 Terra for faster, cheaper computer use(2026-10-27, 46 posts)
- Episode 5: GPT-6.1 Terra Slashes Factual Error Rate from 22.3% to 4.3%(2026-10-27, 3 posts)
- Episode 6: Artificial Analysis Rolls Out Intelligence Index v4.3.2 with GPT-6.1 Terra(2026-10-27, 3 posts)
- Episode 7: OpenAI Quietly Swaps Pro Default Model and Halves Usage Quota(2026-10-27, 2 posts)
- Episode 8: GPT-6.1 Terra Questioned as Benchmark Ties with Sol but Costs More(2026-10-27, 3 posts)
- GPT-6.1 Terra cuts hallucination rate by 50 points while lifting accuracy 2 points — ArtificialAnlys · 2026-10-27
- GPT-6.1 Terra cuts factual error rate from 22.3% to 4.3%, an ~81% drop — haider1 · 2026-10-27
- GPT 6.1 Terra reportedly cuts factual error rate from 22.3% to 4.3% on internal hallucination bench — JeremyNguyenPhD · 2026-10-27