Model comparison
GPT-6 Sol vs gpt-oss-20b
GPT-6 Sol is the stronger model overall, scoring 61.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 111× less per token, which makes it the better buy when GPT-6 Sol's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. GPT-6 Sol scores higher in 9 categories and gpt-oss-20b in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Sol leads 74.0 to 19.3.
- The biggest single-benchmark swing is LMCA: 59.1% for GPT-6 Sol and 14.5% for gpt-oss-20b.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $2 / $10 for GPT-6 Sol.
- GPT-6 Sol accepts more context: 1.05M tokens versus 131K.
- gpt-oss-20b has downloadable open weights; the other is API-only.
Side by side
| GPT-6 Sol | gpt-oss-20b | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 61.8 | 32.5 |
| Released | 2026-09-22 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 128K | 16K |
| Input $ / M tokens | $2 | $0.018 |
| Output $ / M tokens | $10 | $0.09 |
| Results tracked | 45 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
GPT-6 Sol: 60.1 (#11), gpt-oss-20b: 37.6 (#192)
| Benchmark | GPT-6 Sol | gpt-oss-20b |
|---|---|---|
| SciCode | 57.6% | 34.4% |
| LMArena Coding | 1447 | 1306 |
| ALE-Bench | 2,462 | 566.05 |
| DeepSWE | 68.8% | — |
| FrontierCode | 49.3% | — |
| LMArena WebDev | 1688 | — |
| WeirdML | — | 40.9% |
Agentic & Tool Use GPT-6 Sol leads
GPT-6 Sol: 37.2 (#36), gpt-oss-20b: 9.3 (#154)
| Benchmark | GPT-6 Sol | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
| APEX-Agents | 54.3% | — |
| GDP.pdf | 26.4% | — |
| Vending-Bench 2 | 14,428 | — |
Reasoning GPT-6 Sol leads
GPT-6 Sol: 74.0 (#9), gpt-oss-20b: 19.3 (#261)
| Benchmark | GPT-6 Sol | gpt-oss-20b |
|---|---|---|
| CritPt | 30.9% | 1.4% |
| LMArena Hard Prompts | 1418 | 1274 |
| DTBench | 97.3% | 68% |
| LMCA | 59.1% | 14.5% |
| Epoch Capabilities Index | 162.72 | 137.82 |
| ARC-AGI-2 | 89.6% | — |
| Kagi LLM Benchmark | — | 53.2% |
| NYT Connections (extended) | 90.1% | — |
| ARC-AGI-1 | 95.5% | — |
| Chess Puzzles | — | 4% |
| EBR-Bench | 53.3% | — |
| Mystery Game Puzzles | 56% | — |
Math GPT-6 Sol leads
GPT-6 Sol: 87.2 (#7), gpt-oss-20b: 39.4 (#103)
| Benchmark | GPT-6 Sol | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 65.3% |
| LMArena Math | 1402 | 1317 |
| FrontierMath (Tiers 1-3) | 89.8% | — |
| FrontierMath Tier 4 | 90% | — |
| ProofBench | 83% | — |
| Omni-MATH | — | 56.5% |
Knowledge GPT-6 Sol leads
GPT-6 Sol: 64.8 (#15), gpt-oss-20b: 34.6 (#195)
| Benchmark | GPT-6 Sol | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 94.3% | 60.8% |
| LMArena Expert | 1439 | 1258 |
| SimpleQA Verified | 60.7% | — |
| MMLU-Pro | — | 74% |
| Vectara Hallucination Rate | 6.5% | — |
| GPQA (HELM) | — | 59.4% |
Multimodal Not comparable
GPT-6 Sol: 47.6 (#10), gpt-oss-20b: —
| Benchmark | GPT-6 Sol | gpt-oss-20b |
|---|---|---|
| LMArena Vision | 1245 | — |
| Blueprint-Bench 2 | 36.9% | — |
| Furniture Assembly | 58.3% | — |
Multilingual GPT-6 Sol leads
GPT-6 Sol: 50.5 (#118), gpt-oss-20b: 42.2 (#197)
| Benchmark | GPT-6 Sol | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1385 | 1268 |
| LMArena Chinese | 1405 | 1314 |
| LMArena German | 1390 | 1255 |
| LMArena Japanese | 1385 | 1244 |
| LMArena Korean | 1341 | 1236 |
| LMArena Russian | 1401 | 1278 |
| LMArena Spanish | 1384 | 1267 |
| LMArena French | 1410 | — |
Instruction Following GPT-6 Sol leads
GPT-6 Sol: 74.5 (#94), gpt-oss-20b: 61.8 (#240)
| Benchmark | GPT-6 Sol | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1412 | 1236 |
| IFEval | — | 73.2% |
Long Context GPT-6 Sol leads
GPT-6 Sol: 43.1 (#108), gpt-oss-20b: 37.9 (#209)
| Benchmark | GPT-6 Sol | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1411 | 1250 |
Writing & Preference GPT-6 Sol leads
GPT-6 Sol: 71.9 (#18), gpt-oss-20b: 35.5 (#265)
| Benchmark | GPT-6 Sol | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1395 | 1287 |
| LMArena Creative Writing | 1378 | 1201 |
| EQ-Bench Creative Writing | 2125 | 666 |
| LMArena Multi-Turn | 1412 | 1268 |
| WildBench | — | 73.7% |
Frequently asked questions
Is GPT-6 Sol better than gpt-oss-20b?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 111× 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-6 Sol or gpt-oss-20b?
gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; GPT-6 Sol lists at $2 and $10.
Is GPT-6 Sol or gpt-oss-20b better for coding?
GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 37.6 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Sol does, with 1.05M tokens against 131K.
How many benchmarks do GPT-6 Sol and gpt-oss-20b share?
25 benchmarks have published results for both models. GPT-6 Sol has 45 scored results on Noometry and gpt-oss-20b has 34.