GPT-6.1 Sol vs GPT-6.1 Terra
As of Nov 3, 2026, GPT-6.1 Sol leads GPT-6.1 Terra in 4 of 4 shared categories: General LLM, Science, Coding › Web dev, and Legal.
GPT-6.1 Sol
Yes. SOTA in Vision (tied).
GPT-6.1 Terra
#8 Best current rank: #8 in Science, as of Nov 3, 2026.
| Category | GPT-6.1 Sol | GPT-6.1 Terra |
|---|---|---|
| General LLM | #25 · #25 in General LLM Margin 16.603511431097644 Vals AI · Nov 3, 2026 | #67 · #67 in General LLM Margin 37.26790450928383 Vals AI · Nov 3, 2026 |
| Reasoning | #17 · #17 in Reasoning Margin 10.714285714285708 LMArena leaderboard dataset · Oct 28, 2026 | — |
| Science | #4 · #4 in Science Margin — Epoch AI benchmarking hub · Sep 29, 2026 | #8 · #8 in Science Margin — Epoch AI benchmarking hub · Oct 26, 2026 |
| Coding | #3 · Near-SOTA · Coding Margin 12.261503928170598 Terminal-Bench Harbor Hub · Nov 3, 2026 | — |
| Coding › Agentic | #2 · Near-SOTA · Coding › Agentic Margin 13.63636363636364 Terminal-Bench Harbor Hub · Oct 23, 2026 | — |
| Coding › Web dev | #5 · Near-SOTA · Coding › Web dev Margin 3.125 LMArena leaderboard dataset · Oct 28, 2026 | #13 · #13 in Coding › Web dev Margin 9.375 LMArena leaderboard dataset · Oct 28, 2026 |
| Long context | #20 · #20 in Long context Margin 12.5 LMArena leaderboard dataset · Oct 28, 2026 | — |
| Vision | #1 · Co-SOTA · Vision Margin 0 Epoch AI benchmarking hub · Sep 29, 2026 | — |
| Factuality | #34 · #34 in Factuality Margin 20.858895705521476 LMArena leaderboard dataset · Oct 28, 2026 | — |
| Abstraction (ARC) | #1 · #1 in Abstraction (ARC) Margin — Epoch AI benchmarking hub · Sep 29, 2026 | — |
| Cybersecurity | #2 · Near-SOTA · Cybersecurity Margin 3.125 Vals AI · Oct 27, 2026 | — |
| Legal | #20 · #20 in Legal Margin — Vals AI · Nov 3, 2026 | #35 · #35 in Legal Margin — Vals AI · Nov 3, 2026 |
Pricing
GPT-6.1 Sol: $4 per 1M blended tokens.
GPT-6.1 Terra: $2 per 1M blended tokens.
GPT-6.1 Sol: closed weights.
GPT-6.1 Terra: closed weights.
Benchmarks side by side
| Benchmark | GPT-6.1 Sol | GPT-6.1 Terra |
|---|---|---|
| Excel Modeling Benchmark (EMB) | 71.1 | 73.3 |
| Finance Agent | 61.8 | 56.0 |
| GPQA Diamond | 95.4% | 94.3% |
| Legal Research Bench | 38.5 | 26.0 |
| LMArena WebDev | 1751 | 1673 |
| SimpleQA Verified | 73.9% | 57.3% |
| Tax Agent Bench | 21.8 | 17.1 |
| Vals Index | 63.9 | 60.5 |
| Vibe Code Bench | 88.9 | 84.0 |
Latest validated result for each model. Benchmarks where only one model has a result are left out.
Verdict
As of Nov 3, 2026, GPT-6.1 Sol leads GPT-6.1 Terra in 4 of 4 shared categories: General LLM, Science, Coding › Web dev, and Legal.
General LLM: As of Nov 3, 2026, GPT-6.1 Sol leads GPT-6.1 Terra in General LLM at #25 against #67.
Science: As of Nov 3, 2026, GPT-6.1 Sol leads GPT-6.1 Terra in Science at #4 against #8.
Coding › Web dev: As of Nov 3, 2026, GPT-6.1 Sol leads GPT-6.1 Terra in Coding › Web dev at #5 against #13.
Legal: As of Nov 3, 2026, GPT-6.1 Sol leads GPT-6.1 Terra in Legal at #20 against #35.
GPT-6.1 Sol or GPT-6.1 Terra?
As of Nov 3, 2026, GPT-6.1 Sol leads GPT-6.1 Terra in 4 of 4 shared categories: General LLM, Science, Coding › Web dev, and Legal.
GPT-6.1 Sol or GPT-6.1 Terra for General LLM?
As of Nov 3, 2026, GPT-6.1 Sol leads GPT-6.1 Terra in General LLM at #25 against #67.
GPT-6.1 Sol or GPT-6.1 Terra for Science?
As of Nov 3, 2026, GPT-6.1 Sol leads GPT-6.1 Terra in Science at #4 against #8.
GPT-6.1 Sol or GPT-6.1 Terra for Coding › Web dev?
As of Nov 3, 2026, GPT-6.1 Sol leads GPT-6.1 Terra in Coding › Web dev at #5 against #13.