GPT 6.1 Sol outperforms GPT 6.1 Terra at the same cost in nearly every benchmark
Lucky_Creme_5208 · reddit · 2026-10-27
A Reddit user questions whether GPT 6.1 Terra makes sense: on nearly every benchmark except Agents' Last Exam, GPT 6.1 Sol outperforms Terra at almost identical cost, with comparison screenshots attached. Discussion centers on the confusing positioning of same-priced sibling models.
Related event: GPT-6.1 Terra Questioned as Benchmark Ties with Sol but Costs More(3 posts)→
More from Models
- Steering vectors turn Qwen3.5-4B into "Golden Gate Qwen" with a known recipe — burny_tech · 2026-10-27
- Experiment finds 9 'gender neurons' in Qwen3.5-4B that flip its self-identity — burny_tech · 2026-10-27
- MiniMax M3.1 Flash lands on Together AI with LoRA post-training for SFT and RL — MiniMax_AI · 2026-10-27
- Ethan Mollick resurfaces Copilot's classic strawberry fail: 'Two and a half years from this' — aran_nayebi · 2026-10-27
- Claude Opus 5.5 generates explainer video for OpenAI's three discoveries — iScienceLuvr · 2026-10-27
- X forces "Paid Partnership" label on unpaid Kimi K3.1 Pro tweet, locks author's account — RazRazcle · 2026-10-27