GPT-5.6 Sol
| Released | 2026-07-09 |
|---|---|
| Price per 1M | $4.00 in · $20.00 out |
| Overall score | 62.5 · rank 14 |
| Context window | 1,050,000 tokens |
| Provider | OpenAI |
| Weights | Proprietary |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 62.5 | 14 | 54 |
| Overall: Coding | 61.4 | 19 | 39 |
| Overall: Reasoning | 64.2 | 12 | 20 |
| Overall: Agentic work | 62.3 | 16 | 37 |
| Overall: Frontend | 60.0 | 22 | 17 |
| Coding | 61.2 | 18 | 13 |
| Agents | 62.0 | 15 | 11 |
| Reasoning | 65.1 | 11 | 13 |
| Math | 64.3 | 13 | 4 |
| Knowledge | 61.7 | 18 | 3 |
| Frontend | 59.0 | 30 | 2 |
| General | 63.3 | 13 | 6 |
| Multimodal | 61.8 | 8 | 2 |
| Value vs price | +8.5 | 44 |
Buzz
| Posts on X | About 35 a day |
|---|---|
| Mood | 0% positive, 15% negative |
| Hacker News | 6 mentions this week |
| Checked | 2026-11-11 |
- $CBRS Artur Meyster highlights Cerebras’ wafer-scale engineering moat. My additional bull case is that Cerebras could help frontier labs develop their next generation of models faster, accelerating the research process that leads toward recursive self-improvement. RSI means AI …
- CAN AI BUILD BETTER AI? Of the 27 models tested on @ValsAI's AI research eval, only 7 were able to improve the trained model they were given on a fixed compute budget and none of them were open models. Claude Fable 5.2 had the highest score at 37.88, while the best open model w…
- if your AI agent can only read source code, congratulations: you are testing it on training wheels every frontier lab thought autonomous agents had solved software engineering, until Columbia, Google DeepMind, and Anthropic tested them on compiled binaries here is the reality …
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 62.8 ±3.2 | 7 |
| High | 61.5 ±3.4 | 12 |
| Low | 57.4 ±3.7 | 11 |
| Max | 62.5 ±0.9 | 36 |
| Medium | 59.6 ±4.0 | 8 |
| Non-reasoning | 51.8 ±4.2 | 6 |
| Extra High | 61.7 ±1.7 | 15 |
Evidence
Coding 61.2 ±1.4
| Benchmark | Result | Source | Date |
|---|---|---|---|
| CursorBench | 41.7% | Epoch AI | 2026-11-11 |
| FrontierSWE | 32.2% | Epoch AI | 2026-11-11 |
| DeepSWE | 72.7% | Epoch AI | 2026-11-11 |
| FrontierCode | 47.5% | Epoch AI | 2026-11-11 |
| SWE-rebench | 62.3% | SWE-rebench | 2026-07-01 |
| WeirdML v3 | 15.0% | Epoch AI | 2026-11-11 |
| MirrorCode | 20.0% | Epoch AI | 2026-08-10 |
| GSO | 76.5% | Epoch AI | 2026-11-11 |
| IOI | 91.2% | Vals AI | 2026-10-29 |
| SWE-bench Verified (Vals run) | 96.2% | Vals AI | 2026-09-01 |
| SciCode | 57.8% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 82.6% | Vals AI | 2026-09-01 |
| ProgramBench | 1.5% | Vals AI | 2026-11-09 |
Agents 62.0 ±0.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| OSWorld 2.0 | 27.3% | Epoch AI | 2026-11-11 |
| Terminal-Bench 4 | 38.5% | Terminal-Bench | 2026-06-26 |
| SRE Bench | 31.7% | Vals AI | 2026-11-05 |
| Terminal-Bench-Science | 20.0% | Vals AI | 2026-11-09 |
| Terminal-Bench 4.0 (Vals run) | 37.9% | Vals AI | 2026-11-09 |
| Terminal-Bench 2.1 (Vals run) | 85.8% | Vals AI | 2026-10-01 |
| tau2-bench Banking Knowledge | 46.9% | tau2-bench | 2026-07-22 |
| Tax Agent Bench | 31.1% | Vals AI | 2026-11-09 |
| SkillsBench | 54.1% | Vals AI | 2026-10-29 |
| BALROG | 60.0% | Epoch AI | 2026-11-11 |
| Vending-Bench 2 | $9,619 | Epoch AI | 2026-11-11 |
| CyberBench v1.1 | 76.3% | Vals AI | 2026-11-05 |
Reasoning 65.1 ±1.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 92.5% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 92.5% | ARC Prize | 2026-11-11 |
| Extended NYT Connections | 93.8% | Lech Mazur | 2026-11-10 |
| MysteryMechanism | 33.3% | Vals AI | 2026-11-09 |
| CritPt | 32.3% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 97.5% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 97.5% | Epoch AI | 2026-11-11 |
| EnigmaEval | 37.1% | Epoch AI | 2026-11-11 |
| GPQA Diamond | 93.5% | Epoch AI | 2026-07-09 |
| GPQA Diamond (Vals run) | 95.2% | Vals AI | 2026-09-01 |
| BioMysteryBench | 71.1% | Vals AI | 2026-11-09 |
| ARC-AGI-3 | 7.8% | ARC Prize | 2026-11-11 |
| Kagi Reasoning Benchmark | 67.0% | Kagi | 2026-11-11 |
| Chess Puzzles | 59.0% | Epoch AI | 2026-10-13 |
| SimpleBench | 64.8% | SimpleBench | 2026-07-09 |
| SimpleBench | 64.8% | Epoch AI | 2026-11-11 |
| Mystery Game Puzzles | 58.0% | Epoch AI | 2026-07-28 |
Math 64.3 ±1.5
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 | 89.1% | Epoch AI | 2026-07-09 |
| FrontierMath Tier 4 | 82.9% | Epoch AI | 2026-07-09 |
| OTIS Mock AIME 2024-2025 | 100.0% | Epoch AI | 2026-07-09 |
| ProofBench v1.1 | 83.0% | Vals AI | 2026-11-09 |
| ProofBench | 83.0% | Epoch AI | 2026-11-11 |
Knowledge 61.7 ±3.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| SimpleQA Verified | 69.7% | Epoch AI | 2026-08-10 |
| MMLU-Pro (Vals run) | 89.1% | Vals AI | 2026-09-01 |
| Legal Research Bench | 48.1% | Vals AI | 2026-11-09 |
Frontend 59.0 ±0.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Vibe Code Bench v1.1 | 80.5% | Vals AI | 2026-11-09 |
| LMArena WebDev | 1,615 | LMArena | 2026-11-09 |
General 63.3 ±1.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LLM Debate Benchmark | 1,637 | Lech Mazur | 2026-10-26 |
| EQ-Bench 4 | 1,236 | EQ-Bench | 2026-10-24 |
| EQ-Bench Creative Writing v3 | 1,974 | EQ-Bench | 2026-11-11 |
| LMArena Text | 1,485 | LMArena | 2026-11-09 |
| Creative Story-Writing (Lech Mazur) | 82.0% | Lech Mazur | 2026-10-23 |
| EQ-Bench Longform Creative Writing | 81.7% | EQ-Bench | 2026-11-11 |
Multimodal 61.8 ±2.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMMU-Pro (Vals run) | 88.8% | Vals AI | 2026-09-01 |
| LMArena Vision | 1,288 | LMArena | 2026-11-09 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| ARC-AGI-1 | 74.5% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 96.5% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 97.0% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 92.5% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 74.5% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 96.5% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 97.0% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 92.5% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 | 42.5% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 | 85.4% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 | 90.0% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 | 67.1% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 42.5% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 85.4% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 90.0% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 67.1% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 0.3% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 2.2% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 7.0% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-3 | 1.1% | ARC Prize | A stronger configuration result was used |
| Chess Puzzles | 7.0% | Epoch AI | A stronger configuration result was used |
| Chess Puzzles | 31.0% | Epoch AI | A stronger configuration result was used |
| Creative Story-Writing (Lech Mazur) | 80.0% | Lech Mazur | A stronger configuration result was used |
| CritPt | 5.1% | Epoch AI | A stronger configuration result was used |
| CritPt | 14.9% | Epoch AI | A stronger configuration result was used |
| CritPt | 25.7% | Epoch AI | A stronger configuration result was used |
| CritPt | 28.6% | Epoch AI | A stronger configuration result was used |
| CritPt | 22.9% | Epoch AI | A stronger configuration result was used |
| CursorBench | 24.6% | Epoch AI | A stronger configuration result was used |
| CursorBench | 35.7% | Epoch AI | A stronger configuration result was used |
| CursorBench | 37.7% | Epoch AI | A stronger configuration result was used |
| CursorBench | 31.1% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 45.4% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 69.4% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 70.7% | Epoch AI | A stronger configuration result was used |
| DeepSWE | 61.1% | Epoch AI | A stronger configuration result was used |
| Extended NYT Connections | 91.0% | Lech Mazur | A stronger configuration result was used |
| GPQA Diamond | 82.8% | Epoch AI | A stronger configuration result was used |
| GPQA Diamond | 89.9% | Epoch AI | A stronger configuration result was used |
| Mystery Game Puzzles | 33.0% | Epoch AI | A stronger configuration result was used |
| Mystery Game Puzzles | 26.0% | Epoch AI | A stronger configuration result was used |
| OTIS Mock AIME 2024-2025 | 68.9% | Epoch AI | A stronger configuration result was used |
| OTIS Mock AIME 2024-2025 | 95.6% | Epoch AI | A stronger configuration result was used |
| SciCode | 47.1% | Epoch AI | A stronger configuration result was used |
| SciCode | 56.4% | Epoch AI | A stronger configuration result was used |
| SciCode | 57.1% | Epoch AI | A stronger configuration result was used |
| SciCode | 57.1% | Epoch AI | A stronger configuration result was used |
| SciCode | 57.4% | Epoch AI | A stronger configuration result was used |
| SimpleBench | 64.8% | Epoch AI | A stronger configuration result was used |
Names matched to this model
gpt-5-6, gpt-5-6-sol, gpt-daybreak-blue, gpt-daybreak-blue-latest