GPT-6 Sol Max
| Released | 2026-09-22 |
|---|---|
| Price per 1M | $2.00 in · $10.00 out |
| Overall score | 61.6 · rank 70 |
| Context window | 1,050,000 tokens |
| Provider | OpenAI |
| Weights | Proprietary |
| Configuration | Max |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 61.6 | 70 | 31 |
| Overall: Coding | 62.1 | 58 | 23 |
| Overall: Reasoning | 63.9 | 52 | 15 |
| Overall: Agentic work | 62.4 | 55 | 21 |
| Overall: Frontend | 61.2 | 55 | 8 |
| Coding | 61.6 | 63 | 5 |
| Agents | 62.1 | 39 | 7 |
| Reasoning | 65.2 | 40 | 9 |
| Math | 64.9 | 20 | 4 |
| Knowledge | 59.4 | 39 | 2 |
| Frontend | 61.8 | 25 | 2 |
| General | 56.5 | 92 | 1 |
| Multimodal | 56.6 | 61 | 1 |
| Value vs price | +7.4 | 81 |
Buzz
| Posts on X | About 22 a day |
|---|---|
| Mood | 10% positive, 10% negative |
| Hacker News | 3 mentions this week |
| Checked | 2026-11-11 |
- @GoogleDeepMind Carbon ties for #4 on Incident-Arena with a 45.7% pass rate. Carbon levels with GPT-5.6 Terra and ahead of Claude Fable 5.1, GPT-6 Sol, and Gemini 4 Argon, but behind GPT-6 Astra, Claude Opus 5.5, and GPT-6.1 Sol. https://t.co/obRXlwmkaA
- No, GPT 6 Sol, we should not import the entire 810-file Rune source tree to our project in order to fix a Rust compatibility issue. Just increase our Rust version.
- Yesterday we showed what 4 AI models did with 3 leads a sales team was already working: Claude Haiku 4.5 and Claude Sonnet 5.5 from @AnthropicAI, and GPT-5.4 Mini and GPT-6 Sol from @OpenAI. Here's the setup that caught it, and how you'd build one for your own agent. A LayerLen…
Evidence
Coding 61.6 ±1.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierCode | 49.3% | Epoch AI | 2026-11-11 |
| MirrorCode | 48.9% | Epoch AI | 2026-10-29 |
| IOI | 82.6% | Vals AI | 2026-10-29 |
| SciCode | 57.6% | Epoch AI | 2026-11-11 |
| ProgramBench | 2.0% | Vals AI | 2026-11-09 |
Agents 62.1 ±1.4
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 4 | 50.3% | Terminal-Bench | 2026-09-22 |
| SRE Bench | 30.5% | Vals AI | 2026-11-05 |
| Terminal-Bench-Science | 30.0% | Vals AI | 2026-11-09 |
| APEX-Agents | 54.3% | Epoch AI | 2026-11-11 |
| Terminal-Bench 4.0 (Vals run) | 44.4% | Vals AI | 2026-11-09 |
| Terminal-Bench 2.1 (Vals run) | 83.1% | Vals AI | 2026-10-01 |
| Tax Agent Bench | 15.0% | Vals AI | 2026-11-09 |
| CyberBench v1.1 | 78.0% | Vals AI | 2026-11-05 |
Reasoning 65.2 ±1.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 89.6% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 89.6% | ARC Prize | 2026-11-11 |
| MysteryMechanism | 30.2% | Vals AI | 2026-11-09 |
| GPQA Diamond | 94.3% | Epoch AI | 2026-09-22 |
| CritPt | 30.9% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 95.5% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 95.5% | Epoch AI | 2026-11-11 |
| BioMysteryBench | 74.8% | Vals AI | 2026-11-09 |
| ARC-AGI-3 | 23.0% | ARC Prize | 2026-11-11 |
| Chess Puzzles | 56.0% | Epoch AI | 2026-10-13 |
| Mystery Game Puzzles | 56.0% | Epoch AI | 2026-09-22 |
Math 64.9 ±1.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 | 89.8% | Epoch AI | 2026-09-22 |
| FrontierMath Tier 4 | 90.0% | Epoch AI | 2026-09-22 |
| ProofBench v1.1 | 83.0% | Vals AI | 2026-11-09 |
| OTIS Mock AIME 2024-2025 | 100.0% | Epoch AI | 2026-09-22 |
Knowledge 59.4 ±2.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SimpleQA Verified | 60.7% | Epoch AI | 2026-09-22 |
| Legal Research Bench | 28.8% | Vals AI | 2026-11-09 |
Frontend 61.8 ±0.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Vibe Code Bench v1.1 | 87.8% | Vals AI | 2026-11-09 |
| LMArena WebDev | 1,674 | LMArena | 2026-11-09 |
General 56.5 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,457 | LMArena | 2026-11-09 |
Multimodal 56.6 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Vision | 1,258 | LMArena | 2026-11-09 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| ARC-AGI-3 | 4.6% | ARC Prize | Lower duplicate run on the same benchmark |
Names matched to this model
gpt-6-sol-max