Model comparison
GPT-6.1 Sol vs Qwen3.8 Max
GPT-6.1 Sol is the stronger model overall, scoring 65.6 to 56.8 on the Noometry Index.
Last verified . 30 shared benchmarks.
Alibaba (Qwen)Summary
- They share 30 benchmarks with published results for both. GPT-6.1 Sol scores higher in 5 categories and Qwen3.8 Max in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6.1 Sol leads 81.9 to 54.4.
- The biggest single-benchmark swing is Furniture Assembly: 80% for GPT-6.1 Sol and 20% for Qwen3.8 Max.
- Qwen3.8 Max is cheaper at $2 / $6 per million input/output tokens, against $2 / $10 for GPT-6.1 Sol.
- GPT-6.1 Sol accepts more context: 1.05M tokens versus 1M.
Side by side
| GPT-6.1 Sol | Qwen3.8 Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 65.6 | 56.8 |
| Released | 2026-09-29 | 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 | 34 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6.1 Sol leads
GPT-6.1 Sol: 63.2 (#8), Qwen3.8 Max: 53.5 (#29)
| Benchmark | GPT-6.1 Sol | Qwen3.8 Max |
|---|---|---|
| DeepSWE | 75.2% | 57.5% |
| LMArena WebDev | 1755 | 1674 |
| SciCode | 55.8% | 53.2% |
| LMArena Coding | 1487 | 1502 |
| FrontierCode | 50.2% | — |
| FrontierSWE | — | 17.8% |
Agentic & Tool Use Qwen3.8 Max leads
GPT-6.1 Sol: 39.6 (#26), Qwen3.8 Max: 45.4 (#14)
| Benchmark | GPT-6.1 Sol | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | 60% | 63.3% |
| GDP.pdf | 32% | 23.2% |
| τ²-bench Banking | — | 55.1% |
Reasoning GPT-6.1 Sol leads
GPT-6.1 Sol: 81.9 (#2), Qwen3.8 Max: 54.4 (#26)
| Benchmark | GPT-6.1 Sol | Qwen3.8 Max |
|---|---|---|
| NYT Connections (extended) | 95.5% | 88.3% |
| CritPt | 31.7% | 20% |
| Chess Puzzles | 61% | 40% |
| LMArena Hard Prompts | 1466 | 1496 |
| Mystery Game Puzzles | 80% | 38% |
| Epoch Capabilities Index | 166.09 | 156.41 |
| ARC-AGI-2 | 94.2% | — |
| ARC-AGI-1 | 98.5% | — |
| EBR-Bench | 54.3% | — |
| DTBench | — | 92% |
| LMCA | — | 46.2% |
Math GPT-6.1 Sol leads
GPT-6.1 Sol: 93.7 (#1), Qwen3.8 Max: 73.2 (#20)
| Benchmark | GPT-6.1 Sol | Qwen3.8 Max |
|---|---|---|
| FrontierMath (Tiers 1-3) | 93.7% | 74.7% |
| FrontierMath Tier 4 | 100% | 46.3% |
| OTIS Mock AIME 2024-2025 | 100% | 100% |
| ProofBench | 99% | 58% |
| LMArena Math | 1464 | 1499 |
Knowledge GPT-6.1 Sol leads
GPT-6.1 Sol: 71.8 (#4), Qwen3.8 Max: 61.7 (#27)
| Benchmark | GPT-6.1 Sol | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 95.4% | 92.7% |
| SimpleQA Verified | 73.9% | 47.3% |
| LMArena Expert | 1502 | 1507 |
Multimodal GPT-6.1 Sol leads
GPT-6.1 Sol: 52.7 (#5), Qwen3.8 Max: 37.2 (#75)
| Benchmark | GPT-6.1 Sol | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | 1288 | 1314 |
| Furniture Assembly | 80% | 20% |
Multilingual Qwen3.8 Max leads
GPT-6.1 Sol: 54.3 (#46), Qwen3.8 Max: 56.7 (#18)
| Benchmark | GPT-6.1 Sol | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1438 | 1472 |
| LMArena Chinese | 1477 | 1538 |
| LMArena Russian | 1455 | 1481 |
| LMArena French | — | 1503 |
| LMArena German | — | 1483 |
| LMArena Japanese | — | 1467 |
| LMArena Korean | — | 1461 |
| LMArena Spanish | — | 1492 |
Instruction Following Too close to call
GPT-6.1 Sol: 77.0 (#29), Qwen3.8 Max: 77.6 (#17)
| Benchmark | GPT-6.1 Sol | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1468 | 1479 |
Long Context Too close to call
GPT-6.1 Sol: 44.9 (#54), Qwen3.8 Max: 45.6 (#31)
| Benchmark | GPT-6.1 Sol | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1465 | 1489 |
Writing & Preference Qwen3.8 Max leads
GPT-6.1 Sol: 63.6 (#63), Qwen3.8 Max: 67.1 (#30)
| Benchmark | GPT-6.1 Sol | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1447 | 1483 |
| LMArena Creative Writing | 1432 | 1479 |
| LMArena Multi-Turn | 1449 | 1489 |
Frequently asked questions
Is GPT-6.1 Sol better than Qwen3.8 Max?
GPT-6.1 Sol is the stronger model overall, scoring 65.6 to 56.8 on the Noometry Index.
Which is cheaper, GPT-6.1 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-6.1 Sol lists at $2 and $10.
Is GPT-6.1 Sol or Qwen3.8 Max better for coding?
GPT-6.1 Sol scores higher on coding benchmarks: 63.2 versus 53.5 in the Noometry coding category.
Which has the bigger context window?
GPT-6.1 Sol does, with 1.05M tokens against 1M.
How many benchmarks do GPT-6.1 Sol and Qwen3.8 Max share?
30 benchmarks have published results for both models. GPT-6.1 Sol has 34 scored results on Noometry and Qwen3.8 Max has 39.