Model comparison
GPT-5 Nano vs GPT-6 Sol
GPT-6 Sol is the stronger model overall, scoring 61.8 to 33.5 on the Noometry Index. GPT-5 Nano costs 29× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
Summary
- They share 32 benchmarks with published results for both. GPT-5 Nano scores higher in 1 category and GPT-6 Sol in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Sol leads 87.2 to 29.4.
- The biggest single-benchmark swing is FrontierMath Tier 4: 2.4% for GPT-5 Nano and 90% for GPT-6 Sol.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $2 / $10 for GPT-6 Sol.
- GPT-6 Sol accepts more context: 1.05M tokens versus 400K.
Side by side
| GPT-5 Nano | GPT-6 Sol | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 33.5 | 61.8 |
| Released | 2025-08-07 | 2026-09-22 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $0.05 | $2 |
| Output $ / M tokens | $0.40 | $10 |
| Results tracked | 49 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
GPT-5 Nano: 33.6 (#254), GPT-6 Sol: 60.1 (#11)
| Benchmark | GPT-5 Nano | GPT-6 Sol |
|---|---|---|
| LMArena Coding | 1351 | 1447 |
| ALE-Bench | 718.67 | 2,462 |
| DeepSWE | — | 68.8% |
| FrontierCode | — | 49.3% |
| SWE-bench Verified (bash only) | 34.8% | — |
| LMArena WebDev | — | 1688 |
| SciCode | — | 57.6% |
| WeirdML | 38.1% | — |
Agentic & Tool Use GPT-6 Sol leads
GPT-5 Nano: 25.8 (#106), GPT-6 Sol: 37.2 (#36)
| Benchmark | GPT-5 Nano | GPT-6 Sol |
|---|---|---|
| Terminal-Bench | 21.8% | — |
| APEX-Agents | — | 54.3% |
| Berkeley Function Calling Leaderboard | 51.5% | — |
| GDP.pdf | — | 26.4% |
| Vending-Bench 2 | — | 14,428 |
Reasoning GPT-6 Sol leads
GPT-5 Nano: 16.3 (#306), GPT-6 Sol: 74.0 (#9)
| Benchmark | GPT-5 Nano | GPT-6 Sol |
|---|---|---|
| ARC-AGI-2 | 2.6% | 89.6% |
| ARC-AGI-1 | 20.7% | 95.5% |
| LMArena Hard Prompts | 1328 | 1418 |
| Mystery Game Puzzles | 9% | 56% |
| DTBench | 62.7% | 97.3% |
| LMCA | 7.9% | 59.1% |
| Epoch Capabilities Index | 139.38 | 162.72 |
| Kagi LLM Benchmark | 62.2% | — |
| NYT Connections (extended) | — | 90.1% |
| CritPt | — | 30.9% |
| Chess Puzzles | 27% | — |
| EBR-Bench | — | 53.3% |
| ForecastBench | 59.1 | — |
Math GPT-6 Sol leads
GPT-5 Nano: 29.4 (#241), GPT-6 Sol: 87.2 (#7)
| Benchmark | GPT-5 Nano | GPT-6 Sol |
|---|---|---|
| FrontierMath (Tiers 1-3) | 20% | 89.8% |
| FrontierMath Tier 4 | 2.4% | 90% |
| OTIS Mock AIME 2024-2025 | 81.1% | 100% |
| ProofBench | 12% | 83% |
| LMArena Math | 1317 | 1402 |
| Omni-MATH | 54.6% | — |
| MATH Level 5 | 95.2% | — |
| FrontierMath (Feb 2025 set) | 8.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-6 Sol leads
GPT-5 Nano: 35.9 (#178), GPT-6 Sol: 64.8 (#15)
| Benchmark | GPT-5 Nano | GPT-6 Sol |
|---|---|---|
| GPQA Diamond | 69.4% | 94.3% |
| SimpleQA Verified | 11.7% | 60.7% |
| Vectara Hallucination Rate | 10.5% | 6.5% |
| LMArena Expert | 1321 | 1439 |
| MMLU-Pro | 77.8% | — |
| GPQA (HELM) | 67.9% | — |
Multimodal GPT-6 Sol leads
GPT-5 Nano: 31.3 (#108), GPT-6 Sol: 47.6 (#10)
| Benchmark | GPT-5 Nano | GPT-6 Sol |
|---|---|---|
| LMArena Vision | 1159 | 1245 |
| VPCT | 37.2% | — |
| Blueprint-Bench 2 | — | 36.9% |
| Furniture Assembly | — | 58.3% |
Multilingual GPT-6 Sol leads
GPT-5 Nano: 45.3 (#172), GPT-6 Sol: 50.5 (#118)
| Benchmark | GPT-5 Nano | GPT-6 Sol |
|---|---|---|
| LMArena Non-English | 1313 | 1385 |
| LMArena Chinese | 1356 | 1405 |
| LMArena German | 1327 | 1390 |
| LMArena Japanese | 1226 | 1385 |
| LMArena Korean | 1269 | 1341 |
| LMArena Russian | 1296 | 1401 |
| LMArena Spanish | 1360 | 1384 |
| LMArena French | — | 1410 |
Instruction Following Too close to call
GPT-5 Nano: 75.0 (#79), GPT-6 Sol: 74.5 (#94)
| Benchmark | GPT-5 Nano | GPT-6 Sol |
|---|---|---|
| LMArena Instruction Following | 1306 | 1412 |
| IFEval | 93.2% | — |
Long Context GPT-6 Sol leads
GPT-5 Nano: 31.3 (#281), GPT-6 Sol: 43.1 (#108)
| Benchmark | GPT-5 Nano | GPT-6 Sol |
|---|---|---|
| LMArena Longer Query | 1312 | 1411 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference GPT-6 Sol leads
GPT-5 Nano: 39.1 (#249), GPT-6 Sol: 71.9 (#18)
| Benchmark | GPT-5 Nano | GPT-6 Sol |
|---|---|---|
| LMArena Text | 1320 | 1395 |
| LMArena Creative Writing | 1249 | 1378 |
| EQ-Bench Creative Writing | 705 | 2125 |
| LMArena Multi-Turn | 1311 | 1412 |
| WildBench | 80.6% | — |
Frequently asked questions
Is GPT-5 Nano better than GPT-6 Sol?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 33.5 on the Noometry Index. GPT-5 Nano costs 29× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.
Which is cheaper, GPT-5 Nano or GPT-6 Sol?
GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; GPT-6 Sol lists at $2 and $10.
Is GPT-5 Nano or GPT-6 Sol better for coding?
GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 33.6 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Sol does, with 1.05M tokens against 400K.
How many benchmarks do GPT-5 Nano and GPT-6 Sol share?
32 benchmarks have published results for both models. GPT-5 Nano has 49 scored results on Noometry and GPT-6 Sol has 45.