Model comparison
GPT-5.2 vs GPT-6 Sol
GPT-5.2 is the stronger model overall, scoring 54.1 to 49.2 on the Noometry Index.
Last verified . 19 shared benchmarks.
Summary
- They share 19 benchmarks with published results for both. GPT-5.2 scores higher in 8 categories and GPT-6 Sol in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.2 leads 59.8 to 35.8.
- The biggest single-benchmark swing is EBR-Bench: 23% for GPT-5.2 and 53.3% for GPT-6 Sol.
- GPT-6 Sol is cheaper at $2 / $10 per million input/output tokens, against $1.75 / $14 for GPT-5.2.
- GPT-6 Sol accepts more context: 1.05M tokens versus 400K.
Side by side
| GPT-5.2 | GPT-6 Sol | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 54.1 | 49.2 |
| Released | 2025-12-11 | 2026-09-22 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $1.75 | $2 |
| Output $ / M tokens | $14 | $10 |
| Results tracked | 67 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
GPT-5.2: 52.2 (#32), GPT-6 Sol: 59.6 (#11)
| Benchmark | GPT-5.2 | GPT-6 Sol |
|---|---|---|
| LMArena WebDev | 1415 | 1692 |
| LMArena Coding | 1447 | 1453 |
| SWE-bench Verified | 73.8% | — |
| DeepSWE | — | 68.8% |
| FrontierCode | — | 49.3% |
| SWE-bench Verified (bash only) | 72.8% | — |
| SWE-bench Multilingual | 66.7% | — |
| GSO | 27.4% | — |
| WeirdML | 72.2% | — |
| ALE-Bench | 1,294 | — |
| AlgoTune | 2.05 | — |
Agentic & Tool Use GPT-5.2 leads
GPT-5.2: 40.0 (#24), GPT-6 Sol: 33.5 (#49)
| Benchmark | GPT-5.2 | GPT-6 Sol |
|---|---|---|
| Terminal-Bench | 64.9% | — |
| Berkeley Function Calling Leaderboard | 55.9% | — |
| GDPval | 49.7% | — |
| Remote Labor Index | 2.5% | — |
| τ²-bench Airline | 83% | — |
| τ²-bench Banking | 32.2% | — |
| τ²-bench Retail | 81.6% | — |
| τ²-bench Telecom | 89.7% | — |
| DeepResearch Bench | 41.1% | — |
| GDP.pdf | — | 28% |
| LMArena Search | 1207 | — |
| METR Time Horizons | 75.3% | — |
| Vending-Bench 2 | 3,591 | — |
Reasoning Too close to call
GPT-5.2: 50.1 (#33), GPT-6 Sol: 49.3 (#34)
| Benchmark | GPT-5.2 | GPT-6 Sol |
|---|---|---|
| NYT Connections (extended) | 83.6% | 90.1% |
| EBR-Bench | 23% | 53.3% |
| LMArena Hard Prompts | 1445 | 1420 |
| ARC-AGI-2 | 52.9% | — |
| SimpleBench | 45.8% | — |
| Kagi LLM Benchmark | 73.3% | — |
| ARC-AGI-1 | 86.2% | — |
| Chess Puzzles | 49% | — |
| EnigmaEval | 10.4% | — |
| Mystery Game Puzzles | 23% | — |
| DTBench | 90.9% | — |
| LMCA | 43.9% | — |
| Epoch Capabilities Index | 153.45 | — |
| ForecastBench | 60.1 | — |
Math GPT-5.2 leads
GPT-5.2: 59.8 (#32), GPT-6 Sol: 35.8 (#135)
| Benchmark | GPT-5.2 | GPT-6 Sol |
|---|---|---|
| LMArena Math | 1439 | 1404 |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 31.7% | — |
| MathArena Final-Answer Competitions | 72% | — |
| OTIS Mock AIME 2024-2025 | 96.1% | — |
| ProofBench | 15% | — |
| FrontierMath (Feb 2025 set) | 40.7% | — |
| FrontierMath Tier 4 (v1) | 18.8% | — |
Knowledge GPT-5.2 leads
GPT-5.2: 59.2 (#29), GPT-6 Sol: 43.0 (#87)
| Benchmark | GPT-5.2 | GPT-6 Sol |
|---|---|---|
| Vectara Hallucination Rate | 8.4% | 6.5% |
| LMArena Expert | 1445 | 1449 |
| GPQA Diamond | 91.4% | — |
| Humanity's Last Exam | 27.8% | — |
| SimpleQA Verified | 37.1% | — |
Multimodal GPT-5.2 leads
GPT-5.2: 50.9 (#7), GPT-6 Sol: 47.6 (#9)
| Benchmark | GPT-5.2 | GPT-6 Sol |
|---|---|---|
| Furniture Assembly | 38.3% | 58.3% |
| LMArena Vision | 1268 | — |
| VPCT | 84% | — |
| LMArena Document | 1405 | — |
Multilingual GPT-5.2 leads
GPT-5.2: 53.6 (#63), GPT-6 Sol: 51.0 (#113)
| Benchmark | GPT-5.2 | GPT-6 Sol |
|---|---|---|
| LMArena Non-English | 1425 | 1389 |
| LMArena Chinese | 1460 | 1416 |
| LMArena Russian | 1440 | 1407 |
| LMArena Spanish | 1433 | 1377 |
| LMArena French | 1456 | — |
| LMArena German | 1448 | — |
| LMArena Japanese | 1421 | — |
| LMArena Korean | 1390 | — |
Instruction Following Too close to call
GPT-5.2: 74.9 (#83), GPT-6 Sol: 74.4 (#97)
| Benchmark | GPT-5.2 | GPT-6 Sol |
|---|---|---|
| LMArena Instruction Following | 1417 | 1406 |
Long Context Too close to call
GPT-5.2: 44.0 (#74), GPT-6 Sol: 43.2 (#100)
| Benchmark | GPT-5.2 | GPT-6 Sol |
|---|---|---|
| LMArena Longer Query | 1428 | 1415 |
| CL-bench | 18.2% | — |
Writing & Preference GPT-6 Sol leads
GPT-5.2: 67.0 (#30), GPT-6 Sol: 72.3 (#15)
| Benchmark | GPT-5.2 | GPT-6 Sol |
|---|---|---|
| LMArena Text | 1439 | 1396 |
| LMArena Creative Writing | 1402 | 1392 |
| EQ-Bench Creative Writing | 1703 | 2125 |
| LMArena Multi-Turn | 1458 | 1405 |
Frequently asked questions
Is GPT-5.2 better than GPT-6 Sol?
GPT-5.2 is the stronger model overall, scoring 54.1 to 49.2 on the Noometry Index.
Which is cheaper, GPT-5.2 or GPT-6 Sol?
GPT-6 Sol is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; GPT-5.2 lists at $1.75 and $14.
Is GPT-5.2 or GPT-6 Sol better for coding?
GPT-6 Sol scores higher on coding benchmarks: 59.6 versus 52.2 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-5.2 and GPT-6 Sol share?
19 benchmarks have published results for both models. GPT-5.2 has 67 scored results on Noometry and GPT-6 Sol has 22.