GPT-5.6 Sol vs o3
Full profiles: GPT-5.6 Sol (OpenAI) and o3 (OpenAI): every benchmark, price and rank.
Updated · Data from 17 sources
As of 29 Oct 2026, GPT-5.6 Sol is ahead of o3 on quality (65.4 vs 61.0 on the BenchLeader Index), but o3 is 2.3× cheaper at $3.50 vs $8.00 per 1M tokens blended. Choose GPT-5.6 Sol for multimodal (66 vs 57), knowledge (64 vs 56), and instruction following (71 vs 65); choose o3 for fast first answers (5.9 s vs 86.2 s) and agents & tools (70 vs 67), where it also streams ~1.5× faster.
Key facts (as of 29 Oct 2026)
- GPT-5.6 Sol scores 65.4 on the BenchLeader Index and o3 scores 61.0 (BenchLeader, 29 Oct 2026).
- GPT-5.6 Sol costs $8.00 per 1M tokens blended (list price, 29 Oct 2026).
- o3 costs $3.50 per 1M tokens blended (list price, 29 Oct 2026).
- o3 streams ~1.5× faster than GPT-5.6 Sol (Artificial Analysis, 7-day median, 29 Oct 2026).
- Of the 12 benchmarks both models have, GPT-5.6 Sol leads 12 (BenchLeader, 29 Oct 2026).
GPT-5.6 Sol benchmarks Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | GPT-5.6 Sol | 66 vs 62 category score | Medium |
| Agents & tool use | o3 | 70 vs 67 category score; Terminal-Bench Hard 37.1% vs 65.9% | Medium |
| Maths | Either | 71 vs 71 category score | Either |
| Reasoning | GPT-5.6 Sol | 65 vs 61 category score; GPQA Diamond (AA) 94.1% vs 82.7% | Medium |
| Knowledge | GPT-5.6 Sol | 64 vs 56 category score; PRBench Finance 50.5% vs 47.7% | High |
| Long documents | GPT-5.6 Sol | 67 vs 62 category score; AA-LCR v1.1 84.0% vs 74.7% | Medium |
| Images and other media | GPT-5.6 Sol | 66 vs 57 category score; MMMU-Pro 83.4% vs 70.1% | High |
| Following instructions | GPT-5.6 Sol | 71 vs 65 category score; IFBench 72.7% vs 71.4% | Medium |
| Chat quality (human preference) | – | Only o3 has an Arena rating (61 category score) | Not comparable |
| Lowest cost | o3 | $3.50 vs $8.00 per 1M blended; agentic coding session $0.152 vs $0.320 | – |
| Fastest first answer | o3 | 5.9 s (o3) vs 86.2 s (GPT-5.6 Sol (max)) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- GPT-5.6 Sol (max) leads on quality: 65.4 vs 61.0.
- GPT-5.6 Sol (max) is stronger in coding, instruction following, knowledge, long context, maths, multimodal, reasoning.
- o3 is stronger in agents & tools, human preference.
- o3 is 2.3× cheaper ($3.50 vs $8.00 per 1M blended).
- o3 streams 1.9× faster (158 vs 85 tokens per second).
At a glance
| Spec | GPT-5.6 Sol | o3 |
|---|---|---|
| Developer | OpenAI | OpenAI |
| Released | 9 Jul 2026 | 16 Apr 2025 |
| Context window (tokens) | 1,050,000 | 200,000 |
| Compared at | GPT-5.6 Sol (max) | o3 |
| Open weights | No | No |
| Price in / out / cached input per 1M | $4.00 / $20.00 / $0.400 | $2.00 / $8.00 / $0.500 |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | GPT-5.6 Sol (max) | o3 | Leader |
|---|---|---|---|
| Reasoning | 65.1 | 60.9 | GPT-5.6 Sol |
| Coding | 65.8 | 61.8 | GPT-5.6 Sol |
| Agents & tools | 66.6 | 70.3 | o3 |
| Maths | 70.8 | 70.6 | level |
| Knowledge | 63.9 | 56.1 | GPT-5.6 Sol |
| Instruction following | 71.1 | 65.1 | GPT-5.6 Sol |
| Human preference | – | 61.2 | not comparable |
| Multimodal | 66.2 | 56.7 | GPT-5.6 Sol |
| Long context | 66.9 | 62.1 | GPT-5.6 Sol |
Where sources disagree
Across the 12 benchmarks both have, GPT-5.6 Sol leads 12, o3 leads 0. Every overall index that rates both puts GPT-5.6 Sol ahead.
| Overall index | GPT-5.6 Sol (max) | o3 | Leader |
|---|---|---|---|
| BenchLeader Index | 65.4 | 61.0 | GPT-5.6 Sol |
| Arena Text | 1485† | 1432 | GPT-5.6 Sol |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.
| Measure | GPT-5.6 Sol (max) | o3 |
|---|---|---|
| Blended price per 1M tokens | $8.00 | $3.50 |
| Input / output per 1M | $4.00 / $20.00 | $2.00 / $8.00 |
| Cached input per 1M | $0.400 | $0.500 |
| Price per task: chat reply | $0.0076 | $0.0032 |
| Price per task: summarize a 30-page report | $0.060 | $0.029 |
| Price per task: code edit | $0.054 | $0.024 |
| Price per task: agentic coding session | $0.320 | $0.152 |
| Price per task: structured extraction | $0.012 | $0.0056 |
| Output speed | 85 tok/s | 158 tok/s |
| Time to first answer | 86.2 s | 5.9 s |
| Time to a complete answer | 92.1 s | 9.1 s |
Each model is compared at its best-ranked reasoning-effort setting. GPT-5.6 Sol (max) scores 65.4 and takes 86 s to its first answer token; GPT-5.6 Sol (xhigh) scores 64.4 in 22 s, GPT-5.6 Sol (high) scores 63.6 in 11 s. o3 scores 61.0 and takes 6 s to its first answer token; o3 (high) scores 53.8 in 6 s, o3 (medium) scores 53.3 in 6 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
168 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | GPT-5.6 Sol (max) | o3 |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 65.4 | 61.0 |
| Agents & tools score | BenchLeader | 66.6 | 70.3 |
| Coding score | BenchLeader | 65.8 | 61.8 |
| Instruction following score | BenchLeader | 71.1 | 65.1 |
| Knowledge score | BenchLeader | 63.9 | 56.1 |
| Long context score | BenchLeader | 66.9 | 62.1 |
| Maths score | BenchLeader | 70.8 | 70.6 |
| Multimodal score | BenchLeader | 66.2 | 56.7 |
| Reasoning score | BenchLeader | 65.1 | 60.9 |
| Human preference score | BenchLeader | – | 61.2 |
| Blended price $/M | BenchLeader | $8.00 | $3.50 |
| Output speed | BenchLeader | 85 tok/s | 158 tok/s |
| Time to first answer | BenchLeader | 86.2 s | 5.9 s |
| Context window | BenchLeader | 1,050,000 tokens | 200,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 93.5% | 81.8%† |
| Humanity's Last Exam | Scale AI / CAIS | – | 20.3%† |
| SimpleBench | SimpleBench | 64.8%† | 53.1%† |
| Arena Hard Prompts | Arena (LMArena) | 1512† | 1442 |
| LiveBench 1.0 Reasoning | LiveBench | 91.7% | – |
| GPQA Diamond (AA) | Artificial Analysis | 94.1% | 82.7% |
| Humanity's Last Exam (AA) | Artificial Analysis | 49.5% | 20.1% |
| GPQA Diamond (Vals) | Vals AI | 95.2% | 84.1%† |
| EnigmaEval | Scale AI SEAL | 37.1%† | – |
| Kagi LLM Benchmark | Kagi LLM Benchmark | 67.0%† | 67.6% |
| GPQA Diamond (HELM) | HELM Capabilities | – | 75.3% |
| ARC-AGI-1 | ARC Prize | 96.5% | 60.8%† |
| ARC-AGI-2 | ARC Prize | 92.5% | 6.5%† |
| ARC-AGI-3 | ARC Prize | 7.8% | – |
| CritPt | Artificial Analysis | 32.3% | 1.1% |
| CritPt v2 hard subset (AA) | Artificial Analysis | 56.0% | – |
| MysteryMechanism | Vals AI | 33.3% | – |
| SciPredict | Scale AI SEAL | – | 17.9% |
| Chess Puzzles | Epoch AI Benchmarking Hub | 59.0% | 36.0%† |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 58.0% | 29.0%† |
| CL-bench | Epoch AI Benchmarking Hub | – | 17.8%† |
| LMCA | Epoch AI Benchmarking Hub | 58.4% | 39.7%† |
| DTBench | Epoch AI Benchmarking Hub | 95.5% | 84.8%† |
| ForecastBench | ForecastBench | 65.5% | 64.6% |
| BTF-3 | Epoch AI Benchmarking Hub | 13.7% | – |
Show all 168 benchmarks
| Metric | Publisher | GPT-5.6 Sol (max) | o3 |
|---|---|---|---|
| Coding | |||
| SWE-bench Verified (Epoch) | Epoch AI Benchmarking Hub | – | 62.3%† |
| SciCode | SciCode | 57.1% | – |
| WeirdML | WeirdML | 87.0% | 52.4%† |
| FrontierCode | Cognition | 47.5%† | – |
| GSO-Bench | GSO-Bench | 76.5%† | 8.8%† |
| Arena Coding | Arena (LMArena) | 1534† | 1460 |
| Arena WebDev | Arena (LMArena) | 1616† | – |
| LiveBench 1.0 Coding | LiveBench | 83.9% | – |
| LiveCodeBench | Vals AI | 82.6% | 83.9%† |
| IOI | Vals AI | 91.2% | – |
| SWE-bench (Vals) | Vals AI | 96.2% | – |
| Aider Polyglot | Aider polyglot leaderboard | – | 81.3% |
| SWE-bench Verified (bash only) | SWE-bench | – | 58.4% |
| SWE-bench Verified (any scaffold) | SWE-bench | – | 58.4% |
| Code Migration | Vals AI | 52.9% | – |
| ProgramBench | Vals AI | 1.5% | – |
| Vibe Code Bench 1-100 | Vals AI | 20.0% | – |
| Vibe Code Bench v1.1 | Vals AI | 80.5% | – |
| SWE Atlas: Codebase QnA | Scale AI SEAL | 46.0%† | – |
| SWE Atlas: Test Writing | Scale AI SEAL | 45.9%† | – |
| MirrorCode | Epoch AI Benchmarking Hub | 20.0%† | – |
| CadEval | Epoch AI Benchmarking Hub | – | 74.0%† |
| Surface Evolver Bench | Epoch AI Benchmarking Hub | 93.1%† | – |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | 72.7% | – |
| CursorBench | Cursor | 41.7% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 2176.9 | 933.5† |
| FrontierSWE | Epoch AI Benchmarking Hub | 32.2% | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 39.9% | – |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 88.0% | – |
| Startrise Frontend | Startrise | 66.2%† | – |
| LiveCodeBench (AA) | Artificial Analysis | – | 80.8% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 37.3% | – |
| OSWorld-Verified 2.0 | OSWorld | 27.3% | – |
| GDPval | OpenAI | – | 30.8%† |
| Arena Agent | Arena (LMArena) | 6.1† | – |
| LiveBench Agentic Coding | LiveBench | 56.2% | – |
| Terminal-Bench Hard | Artificial Analysis | 65.9% | 37.1% |
| τ²-Bench Telecom (AA) | Artificial Analysis | 85.1% | 80.7% |
| Terminal-Bench 2.1 (Vals) | Vals AI | 85.8% | – |
| HiL-Bench | Scale AI SEAL | 32.3%† | – |
| BFCL Overall | Berkeley Function Calling Leaderboard | – | 63.0% |
| τ³-bench Banking | τ²-bench | 46.9%† | – |
| τ³-Banking (AA) | Artificial Analysis | 44.3% | – |
| ITBench SRE (AA) | Artificial Analysis | 56.2% | – |
| Analyst Agent (AA) | Artificial Analysis | 47.5% | – |
| CUA-bench | Vals AI | 8.3% | – |
| CyberBench | Vals AI | 76.3% | – |
| Finance Agent v2 | Vals AI | 53.8% | – |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 2.5% | – |
| Legal Research Bench | Vals AI | 48.1% | – |
| SkillsBench | Vals AI | 54.1% | – |
| SREBench | Vals AI | 30.5% | – |
| Tax Agent Bench | Vals AI | 31.1% | – |
| Terminal-Bench 4.0 (Vals) | Vals AI | 37.9% | – |
| Terminal-Bench Science | Vals AI | 20.0% | – |
| Time Horizon Index: KSP | Vals AI | 23.8% | – |
| Web Search Index | Vals AI | 43.6% | – |
| Arena Search | Arena (LMArena) | 1258† | – |
| BALROG | BALROG | 60.0% | – |
| DeepResearch Bench | DeepResearch Bench | – | 45.2%† |
| PostTrainBench | Epoch AI Benchmarking Hub | 36.2% | – |
| METR Time Horizons | METR | – | 63.6% |
| GBAEval | Epoch AI Benchmarking Hub | 52.6%† | – |
| Vending-Bench 2 | Andon Labs | 9619.4† | – |
| AutomationBench | Artificial Analysis | 60.1% | – |
| GDP.pdf (AA) | Artificial Analysis | 27.2% | – |
| EnterpriseOps-Gym | Artificial Analysis | 42.9% | – |
| AA-Briefcase v1.1 | Artificial Analysis | 1478 | – |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 22.4% | – |
| BrowseComp-Plus | BrowseComp-Plus | – | 50.5% |
| FACTS Search | 88.5%† | – | |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 89.1% | 33.3%† |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 82.9% | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 100.0% | 83.9%† |
| MATH Level 5 | Epoch AI Benchmarking Hub | – | 97.8%† |
| ProofBench | Vals AI | 83.0% | – |
| LiveBench Mathematics | LiveBench | 96.2% | – |
| AIME (Vals) | Vals AI | – | 85.3%† |
| MGSM | Vals AI | – | 91.8%† |
| IMO 2025 | MathArena | – | 16.7%† |
| Omni-MATH (HELM) | HELM Capabilities | – | 71.4% |
| MATH 500 | Vals AI | – | 94.6%† |
| Arena Maths | Arena (LMArena) | 1496† | 1448 |
| AIME 2025 (AA) | Artificial Analysis | – | 88.3% |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 69.7% | 49.4%† |
| LiveBench 1.0 Data Analysis | LiveBench | 79.8% | – |
| MMLU-Pro | Vals AI | 89.1% | 85.6%† |
| LegalBench | Vals AI | 87.0% | 83.8%† |
| CorpFin | Vals AI | 64.4% | 59.7%† |
| TaxEval | Vals AI | 74.8% | 74.6%† |
| MedQA | Vals AI | – | 96.1%† |
| PRBench Finance | Scale AI SEAL | 50.5% | 47.7% |
| PRBench Legal | Scale AI SEAL | 50.5% | 48.6% |
| MultiNRC | Scale AI SEAL | – | 45.5%† |
| MMLU-Pro (HELM) | HELM Capabilities | – | 85.9% |
| BioMysteryBench | Vals AI | 71.1% | – |
| Excel Modeling Benchmark | Vals AI | 72.3% | – |
| MedCode | Vals AI | 44.0% | 47.3%† |
| MedScribe | Vals AI | 85.2% | 76.7%† |
| Public Benefits Bench v1.1 | Vals AI | 66.5% | – |
| DrugDiscoveryBench | Scale AI SEAL | 56.5% | – |
| EBR-bench | Epoch AI Benchmarking Hub | 44.8% | – |
| GDP.pdf | Epoch AI Benchmarking Hub | 30.7% | – |
| Instruction following | |||
| LiveBench Language | LiveBench | 87.7% | – |
| LiveBench Instruction Following | LiveBench | 71.8% | – |
| IFBench | Artificial Analysis | 72.7% | 71.4% |
| MultiChallenge | Scale AI SEAL | – | 56.6%† |
| TutorBench | Scale AI SEAL | – | 52.1%† |
| IFEval (HELM) | HELM Capabilities | – | 86.9% |
| WildBench (HELM) | HELM Capabilities | – | 86.1% |
| Arena Instruction Following | Arena (LMArena) | 1491† | 1403 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1485† | 1432 |
| EQ-Bench 4 | EQ-Bench | 1236† | – |
| Arena Creative Writing | Arena (LMArena) | 1470† | 1383 |
| Arena Multi-turn | Arena (LMArena) | 1490† | 1420 |
| Arena Longer Queries | Arena (LMArena) | 1499† | 1411 |
| Lech Mazur Creative Writing | Lech Mazur | – | 83.9%† |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1281† | 1214 |
| MMMU-Pro | Artificial Analysis | 83.4% | 70.1% |
| VISTA | Scale AI SEAL | – | 50.1%† |
| MMMU (validation) | MMMU | – | 82.9% |
| MMMU-Pro (official) | MMMU | – | 76.4% |
| MMMU-Pro (Vals) | Vals AI | 88.8% | 80.4%† |
| MortgageTax | Vals AI | 67.3% | 65.7%† |
| VTB | Scale AI SEAL | – | 13.7% |
| GeoBench | GeoBench | – | 60.0%† |
| VPCT | Epoch AI Benchmarking Hub | – | 52.0%† |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 33.6%† | – |
| Long context | |||
| Fiction.LiveBench 120k | Fiction.live | – | 100.0%† |
| AA-LCR v1.1 | Artificial Analysis | 84.0% | 74.7% |
| Arena Document | Arena (LMArena) | 1489† | – |
| MLCR | Artificial Analysis | 30.0% | – |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 162.8† | 146.7 |
| LiveBench 1.0 | LiveBench | 81.0% | – |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 47.0 | 20.2 |
| AA Progress Index v1 | Artificial Analysis | 331.5 | 258.3 |
| HELM Capabilities mean | HELM Capabilities | – | 81.1% |
| Vals RSI Index | Vals AI | 23.9% | – |
| Vals Science Index | Vals AI | 41.5% | – |
| Vals Multimodal Index | Vals AI | 72.6% | – |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 20.1%† | 16.0% |
| MASK | Scale AI SEAL | – | 84.5%† |
| PropensityBench | Scale AI SEAL | – | 10.5% |
Data as of 2026-10-29. Best configuration of each model; every score links to its source on the model pages.
What changed
- GPT-5.6 Sol (max): 2 new results (AA Progress Index v1, CritPt v2 hard subset (AA)).
- o3: 1 new result (AA Progress Index v1).
- GPT-5.6 Sol (max): 1 new result (AA-Briefcase v1.1).
- GPT-5.6 Sol (max): 3 new results (Harvey's Legal Agent Benchmark v1, Vals RSI Index, Vals Science Index).
- GPT-5.6 Sol (max): 1 score revision.
- GPT-5.6 Sol (max): 1 new result (ForecastBench).
- o3: 1 score revision.
- GPT-5.6 Sol (max): 5 new results (AutomationBench, GDP.pdf (AA), MLCR…).
- GPT-5.6 Sol (max): 1 score revision.
- GPT-5.6 Sol (max): 1 score revision.
Related comparisons
- GPT-5.6 Sol vs Claude Opus 5.5
- GPT-5.6 Sol vs GPT-6 Astra
- GPT-5.6 Sol vs GPT-6.1 Sol
- o3 vs Claude Opus 5.5
- o3 vs GPT-6 Astra
- o3 vs GPT-6.1 Sol
- o3 vs Claude Fable 5.1
- GPT-5.5 vs o3
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
GPT-5.6 Sol vs o3: questions
- Is GPT-5.6 Sol better than o3?
- GPT-5.6 Sol (max) leads on quality: 65.4 vs 61.0. The BenchLeader Index combines every independent quality benchmark; GPT-5.6 Sol (max) is ahead overall as of 29 Oct 2026, but check the category scores for your use.
- GPT-5.6 Sol vs o3: which is cheaper, and will that change?
- o3 is cheaper today: $3.50 against $8.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is GPT-5.6 Sol or o3 better for coding?
- GPT-5.6 Sol scores higher in coding (66 vs 62 on the category index, where 50 is average).
- Which is faster, GPT-5.6 Sol or o3?
- o3 streams ~1.5× faster (158 vs 85 output tokens per second, Artificial Analysis 7-day median).
- Is GPT-5.6 Sol or o3 better for agents and tool use?
- o3 scores higher in agentic tasks (70 vs 67 on the category index, where 50 is average). On Terminal-Bench Hard, the top shared agents & tools benchmark, it is 65.9% for GPT-5.6 Sol against 37.1% for o3.
More questions (4)
- Which is better for long documents?
- GPT-5.6 Sol scores higher on long-context benchmarks (67 vs 62). GPT-5.6 Sol accepts more context: 1,050,000 against 200,000 tokens for o3.
- Which has lower latency?
- o3 answers first: 5.9 s against 86.2 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on GPT-5.6 Sol vs o3?
- Across the 12 benchmarks both have, GPT-5.6 Sol leads 12, o3 leads 0. Every overall index that rates both puts GPT-5.6 Sol ahead.
- When was GPT-5.6 Sol released?
- GPT-5.6 Sol was released on 9 Jul 2026.