Model comparison
GPT-6 Sol vs Qwen3.8 Max
GPT-6 Sol is the stronger model overall, scoring 61.8 to 56.8 on the Noometry Index.
Last verified . 36 shared benchmarks.
Alibaba (Qwen)Summary
- They share 36 benchmarks with published results for both. GPT-6 Sol scores higher in 6 categories and Qwen3.8 Max in 4 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Sol leads 74.0 to 54.4.
- The biggest single-benchmark swing is FrontierMath Tier 4: 90% for GPT-6 Sol and 46.3% for Qwen3.8 Max.
- Qwen3.8 Max is cheaper at $2 / $6 per million input/output tokens, against $2 / $10 for GPT-6 Sol.
- GPT-6 Sol accepts more context: 1.05M tokens versus 1M.
Side by side
| GPT-6 Sol | Qwen3.8 Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 61.8 | 56.8 |
| Released | 2026-09-22 | 2026-08-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1M |
| Max output | 128K | 131K |
| Input $ / M tokens | $2 | $2 |
| Output $ / M tokens | $10 | $6 |
| Results tracked | 45 | 39 |
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), Qwen3.8 Max: 53.5 (#29)
| Benchmark | GPT-6 Sol | Qwen3.8 Max |
|---|---|---|
| DeepSWE | 68.8% | 57.5% |
| LMArena WebDev | 1688 | 1674 |
| SciCode | 57.6% | 53.2% |
| LMArena Coding | 1447 | 1502 |
| FrontierCode | 49.3% | — |
| FrontierSWE | — | 17.8% |
| ALE-Bench | 2,462 | — |
Agentic & Tool Use Qwen3.8 Max leads
GPT-6 Sol: 37.2 (#36), Qwen3.8 Max: 45.4 (#14)
| Benchmark | GPT-6 Sol | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | 54.3% | 63.3% |
| GDP.pdf | 26.4% | 23.2% |
| τ²-bench Banking | — | 55.1% |
| Vending-Bench 2 | 14,428 | — |
Reasoning GPT-6 Sol leads
GPT-6 Sol: 74.0 (#9), Qwen3.8 Max: 54.4 (#26)
| Benchmark | GPT-6 Sol | Qwen3.8 Max |
|---|---|---|
| NYT Connections (extended) | 90.1% | 88.3% |
| CritPt | 30.9% | 20% |
| LMArena Hard Prompts | 1418 | 1496 |
| Mystery Game Puzzles | 56% | 38% |
| DTBench | 97.3% | 92% |
| LMCA | 59.1% | 46.2% |
| Epoch Capabilities Index | 162.72 | 156.41 |
| ARC-AGI-2 | 89.6% | — |
| ARC-AGI-1 | 95.5% | — |
| Chess Puzzles | — | 40% |
| EBR-Bench | 53.3% | — |
Math GPT-6 Sol leads
GPT-6 Sol: 87.2 (#7), Qwen3.8 Max: 73.2 (#20)
| Benchmark | GPT-6 Sol | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 89.8% | 74.7% |
| FrontierMath Tier 4 | 90% | 46.3% |
| OTIS Mock AIME 2024-2025 | 100% | 100% |
| ProofBench | 83% | 58% |
| LMArena Math | 1402 | 1499 |
Knowledge GPT-6 Sol leads
GPT-6 Sol: 64.8 (#15), Qwen3.8 Max: 61.7 (#27)
| Benchmark | GPT-6 Sol | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 94.3% | 92.7% |
| SimpleQA Verified | 60.7% | 47.3% |
| LMArena Expert | 1439 | 1507 |
| Vectara Hallucination Rate | 6.5% | — |
Multimodal GPT-6 Sol leads
GPT-6 Sol: 47.6 (#10), Qwen3.8 Max: 37.2 (#75)
| Benchmark | GPT-6 Sol | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | 1245 | 1314 |
| Furniture Assembly | 58.3% | 20% |
| Blueprint-Bench 2 | 36.9% | — |
Multilingual Qwen3.8 Max leads
GPT-6 Sol: 50.5 (#118), Qwen3.8 Max: 56.7 (#18)
| Benchmark | GPT-6 Sol | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1385 | 1472 |
| LMArena Chinese | 1405 | 1538 |
| LMArena French | 1410 | 1503 |
| LMArena German | 1390 | 1483 |
| LMArena Japanese | 1385 | 1467 |
| LMArena Korean | 1341 | 1461 |
| LMArena Russian | 1401 | 1481 |
| LMArena Spanish | 1384 | 1492 |
Instruction Following Qwen3.8 Max leads
GPT-6 Sol: 74.5 (#94), Qwen3.8 Max: 77.6 (#17)
| Benchmark | GPT-6 Sol | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1412 | 1479 |
Long Context Qwen3.8 Max leads
GPT-6 Sol: 43.1 (#108), Qwen3.8 Max: 45.6 (#31)
| Benchmark | GPT-6 Sol | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1411 | 1489 |
Writing & Preference GPT-6 Sol leads
GPT-6 Sol: 71.9 (#18), Qwen3.8 Max: 67.1 (#30)
| Benchmark | GPT-6 Sol | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1395 | 1483 |
| LMArena Creative Writing | 1378 | 1479 |
| LMArena Multi-Turn | 1412 | 1489 |
| EQ-Bench Creative Writing | 2125 | — |
Frequently asked questions
Is GPT-6 Sol better than Qwen3.8 Max?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 56.8 on the Noometry Index.
Which is cheaper, GPT-6 Sol or Qwen3.8 Max?
Qwen3.8 Max is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; GPT-6 Sol lists at $2 and $10.
Is GPT-6 Sol or Qwen3.8 Max better for coding?
GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 53.5 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Sol does, with 1.05M tokens against 1M.
How many benchmarks do GPT-6 Sol and Qwen3.8 Max share?
36 benchmarks have published results for both models. GPT-6 Sol has 45 scored results on Noometry and Qwen3.8 Max has 39.