Model comparison
GPT-5 Mini vs Qwen2.5-Max
GPT-5 Mini is the stronger model overall, scoring 41.7 to 40.1 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. GPT-5 Mini scores higher in 5 categories and Qwen2.5-Max in 3 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.3 to 34.3.
- The biggest single-benchmark swing is Short-Story Creative Writing: 83.1% for GPT-5 Mini and 72.9% for Qwen2.5-Max.
Side by side
| GPT-5 Mini | Qwen2.5-Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 41.7 | 40.1 |
| Released | 2025-08-07 | 2025-01-25 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $2 | — |
| Results tracked | 60 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5-Max leads
GPT-5 Mini: 40.0 (#134), Qwen2.5-Max: 41.2 (#118)
| Benchmark | GPT-5 Mini | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1406 | 1359 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| SWE-bench Multilingual | 39.7% | — |
| SciCode | 39.2% | — |
| WeirdML | 52.7% | — |
| LiveBench Coding | — | 64.4% |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |
Agentic & Tool Use Not comparable
GPT-5 Mini: 31.0 (#64), Qwen2.5-Max: —
| Benchmark | GPT-5 Mini | Qwen2.5-Max |
|---|---|---|
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| Vending-Bench 2 | -31.18 | — |
Reasoning Too close to call
GPT-5 Mini: 24.1 (#153), Qwen2.5-Max: 24.6 (#144)
| Benchmark | GPT-5 Mini | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1380 | 1360 |
| Epoch Capabilities Index | 145.51 | 132.54 |
| ARC-AGI-2 | 4.4% | — |
| Kagi LLM Benchmark | 70.3% | — |
| ARC-AGI-1 | 54.3% | — |
| CritPt | 0% | — |
| Chess Puzzles | 30% | — |
| EnigmaEval | 8.2% | — |
| LiveBench Reasoning | — | 51.4% |
| Mystery Game Puzzles | 10% | — |
| DTBench | 80.5% | — |
| LiveBench Data Analysis | — | 67.9% |
| LMCA | 34.2% | — |
| ForecastBench | 61 | — |
| LiveBench | — | 62.3% |
Math GPT-5 Mini leads
GPT-5 Mini: 46.3 (#59), Qwen2.5-Max: 34.3 (#162)
| Benchmark | GPT-5 Mini | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1377 | 1369 |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| OTIS Mock AIME 2024-2025 | 86.7% | — |
| ProofBench | 9% | — |
| Omni-MATH | 72.2% | — |
| LiveBench Math | — | 58.4% |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 27.2% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5 Mini leads
GPT-5 Mini: 45.5 (#80), Qwen2.5-Max: 34.8 (#181)
| Benchmark | GPT-5 Mini | Qwen2.5-Max |
|---|---|---|
| Confabulations | 13.3% | 21.8% |
| LMArena Expert | 1377 | 1337 |
| GPQA Diamond | 75% | — |
| Humanity's Last Exam | 19.4% | — |
| SimpleQA Verified | 21.6% | — |
| MMLU-Pro | 83.5% | — |
| Vectara Hallucination Rate | 12.9% | — |
| GPQA (HELM) | 75.6% | — |
Multimodal Not comparable
GPT-5 Mini: 35.0 (#80), Qwen2.5-Max: —
| Benchmark | GPT-5 Mini | Qwen2.5-Max |
|---|---|---|
| LMArena Vision | 1202 | — |
| VPCT | 40.2% | — |
Multilingual Too close to call
GPT-5 Mini: 49.1 (#131), Qwen2.5-Max: 48.3 (#140)
| Benchmark | GPT-5 Mini | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1363 | 1352 |
| LMArena Chinese | 1387 | 1382 |
| LMArena French | 1385 | 1396 |
| LMArena German | 1366 | 1350 |
| LMArena Japanese | 1342 | 1299 |
| LMArena Korean | 1306 | 1304 |
| LMArena Russian | 1362 | 1353 |
| LMArena Spanish | 1355 | 1377 |
Instruction Following GPT-5 Mini leads
GPT-5 Mini: 76.4 (#43), Qwen2.5-Max: 71.5 (#147)
| Benchmark | GPT-5 Mini | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1357 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
| IFEval | 92.7% | — |
Long Context Too close to call
GPT-5 Mini: 41.9 (#127), Qwen2.5-Max: 41.3 (#137)
| Benchmark | GPT-5 Mini | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1355 | 1358 |
| Fiction.LiveBench | 69.4% | — |
Writing & Preference Too close to call
GPT-5 Mini: 55.4 (#143), Qwen2.5-Max: 55.6 (#141)
| Benchmark | GPT-5 Mini | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1373 | 1367 |
| LMArena Creative Writing | 1325 | 1339 |
| Short-Story Creative Writing | 83.1% | 72.9% |
| LMArena Multi-Turn | 1363 | 1364 |
| EQ-Bench Creative Writing | 1313 | — |
| WildBench | 85.5% | — |
| LiveBench Language | — | 56.3% |
Frequently asked questions
Is GPT-5 Mini better than Qwen2.5-Max?
GPT-5 Mini is the stronger model overall, scoring 41.7 to 40.1 on the Noometry Index.
Is GPT-5 Mini or Qwen2.5-Max better for coding?
Qwen2.5-Max scores higher on coding benchmarks: 41.2 versus 40.0 in the Noometry coding category.
How many benchmarks do GPT-5 Mini and Qwen2.5-Max share?
20 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Qwen2.5-Max has 27.