Model comparison
GPT-6 Sol vs Qwen2.5-VL 72B Instruct
GPT-6 Sol is the stronger model overall, scoring 61.8 to 29.9 on the Noometry Index.
Last verified . 1 shared benchmarks.
Alibaba (Qwen)Summary
- They share 1 benchmark with published results for both. GPT-6 Sol scores higher in 3 categories and Qwen2.5-VL 72B Instruct in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-6 Sol leads 74.0 to 20.7.
- Both cost about the same: $2 input and $10 output per million tokens.
- GPT-6 Sol accepts more context: 1.05M tokens versus 131K.
- Qwen2.5-VL 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-6 Sol | Qwen2.5-VL 72B Instruct | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 61.8 | 29.9 |
| Released | 2026-09-22 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 128K | 8K |
| Input $ / M tokens | $2 | $2.80 |
| Output $ / M tokens | $10 | $8.40 |
| Results tracked | 45 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
GPT-6 Sol: 60.1 (#11), Qwen2.5-VL 72B Instruct: —
| Benchmark | GPT-6 Sol | Qwen2.5-VL 72B Instruct |
|---|---|---|
| DeepSWE | 68.8% | — |
| FrontierCode | 49.3% | — |
| LMArena WebDev | 1688 | — |
| SciCode | 57.6% | — |
| LMArena Coding | 1447 | — |
| ALE-Bench | 2,462 | — |
Agentic & Tool Use GPT-6 Sol leads
GPT-6 Sol: 37.2 (#36), Qwen2.5-VL 72B Instruct: 18.6 (#144)
| Benchmark | GPT-6 Sol | Qwen2.5-VL 72B Instruct |
|---|---|---|
| APEX-Agents | 54.3% | — |
| OSWorld | — | 5% |
| GDP.pdf | 26.4% | — |
| Vending-Bench 2 | 14,428 | — |
Reasoning GPT-6 Sol leads
GPT-6 Sol: 74.0 (#9), Qwen2.5-VL 72B Instruct: 20.7 (#233)
| Benchmark | GPT-6 Sol | Qwen2.5-VL 72B Instruct |
|---|---|---|
| ARC-AGI-2 | 89.6% | — |
| Kagi LLM Benchmark | — | 36% |
| NYT Connections (extended) | 90.1% | — |
| ARC-AGI-1 | 95.5% | — |
| CritPt | 30.9% | — |
| EBR-Bench | 53.3% | — |
| LMArena Hard Prompts | 1418 | — |
| Mystery Game Puzzles | 56% | — |
| DTBench | 97.3% | — |
| LMCA | 59.1% | — |
| Epoch Capabilities Index | 162.72 | — |
Math Not comparable
GPT-6 Sol: 87.2 (#7), Qwen2.5-VL 72B Instruct: —
| Benchmark | GPT-6 Sol | Qwen2.5-VL 72B Instruct |
|---|---|---|
| FrontierMath (Tiers 1-3) | 89.8% | — |
| FrontierMath Tier 4 | 90% | — |
| OTIS Mock AIME 2024-2025 | 100% | — |
| ProofBench | 83% | — |
| LMArena Math | 1402 | — |
Knowledge Not comparable
GPT-6 Sol: 64.8 (#15), Qwen2.5-VL 72B Instruct: —
| Benchmark | GPT-6 Sol | Qwen2.5-VL 72B Instruct |
|---|---|---|
| GPQA Diamond | 94.3% | — |
| SimpleQA Verified | 60.7% | — |
| Vectara Hallucination Rate | 6.5% | — |
| LMArena Expert | 1439 | — |
Multimodal GPT-6 Sol leads
GPT-6 Sol: 47.6 (#10), Qwen2.5-VL 72B Instruct: 33.5 (#97)
| Benchmark | GPT-6 Sol | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Vision | 1245 | 1107 |
| Video-MME | — | 73.5% |
| GeoBench | — | 62% |
| Blueprint-Bench 2 | 36.9% | — |
| Furniture Assembly | 58.3% | — |
| SpatialViz-Bench | — | 33.3% |
Multilingual Not comparable
GPT-6 Sol: 50.5 (#118), Qwen2.5-VL 72B Instruct: —
| Benchmark | GPT-6 Sol | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Non-English | 1385 | — |
| LMArena Chinese | 1405 | — |
| LMArena French | 1410 | — |
| LMArena German | 1390 | — |
| LMArena Japanese | 1385 | — |
| LMArena Korean | 1341 | — |
| LMArena Russian | 1401 | — |
| LMArena Spanish | 1384 | — |
Instruction Following Not comparable
GPT-6 Sol: 74.5 (#94), Qwen2.5-VL 72B Instruct: —
| Benchmark | GPT-6 Sol | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1412 | — |
Long Context Not comparable
GPT-6 Sol: 43.1 (#108), Qwen2.5-VL 72B Instruct: —
| Benchmark | GPT-6 Sol | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1411 | — |
Writing & Preference Not comparable
GPT-6 Sol: 71.9 (#18), Qwen2.5-VL 72B Instruct: —
| Benchmark | GPT-6 Sol | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Text | 1395 | — |
| LMArena Creative Writing | 1378 | — |
| EQ-Bench Creative Writing | 2125 | — |
| LMArena Multi-Turn | 1412 | — |
Frequently asked questions
Is GPT-6 Sol better than Qwen2.5-VL 72B Instruct?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 29.9 on the Noometry Index.
Which is cheaper, GPT-6 Sol or Qwen2.5-VL 72B Instruct?
GPT-6 Sol is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; Qwen2.5-VL 72B Instruct lists at $2.80 and $8.40.
Which has the bigger context window?
GPT-6 Sol does, with 1.05M tokens against 131K.
How many benchmarks do GPT-6 Sol and Qwen2.5-VL 72B Instruct share?
1 benchmark has published results for both models. GPT-6 Sol has 45 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.