Model comparison
GPT-5 Mini vs Qwen2.5 72B Instruct
GPT-5 Mini is the stronger model overall, scoring 41.8 to 31.9 on the Noometry Index.
Last verified . 31 shared benchmarks.
Alibaba (Qwen)Summary
- They share 31 benchmarks with published results for both. GPT-5 Mini scores higher in 9 categories and Qwen2.5 72B Instruct in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.7 to 19.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 86.7% for GPT-5 Mini and 8.1% for Qwen2.5 72B Instruct.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- GPT-5 Mini accepts more context: 400K tokens versus 131K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Mini | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 41.8 | 31.9 |
| Released | 2025-08-07 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | 400K | 131K |
| Max output | 128K | 8K |
| Input $ / M tokens | $0.25 | $1.40 |
| Output $ / M tokens | $2 | $5.60 |
| Results tracked | 60 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
GPT-5 Mini: 40.1 (#146), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | GPT-5 Mini | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 52.7% | 16% |
| LMArena Coding | 1406 | 1292 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| SWE-bench Multilingual | 39.7% | — |
| SciCode | 39.2% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |
Agentic & Tool Use GPT-5 Mini leads
GPT-5 Mini: 31.1 (#70), Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | GPT-5 Mini | Qwen2.5 72B Instruct |
|---|---|---|
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
| Vending-Bench 2 | -31.18 | — |
Reasoning GPT-5 Mini leads
GPT-5 Mini: 23.9 (#168), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | GPT-5 Mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1380 | 1271 |
| DTBench | 80.5% | 62.9% |
| LMCA | 34.2% | 13.4% |
| Epoch Capabilities Index | 145.52 | 129 |
| ForecastBench | 61 | 57.5 |
| ARC-AGI-2 | 4.4% | — |
| Kagi LLM Benchmark | 70.3% | — |
| ARC-AGI-1 | 54.3% | — |
| CritPt | 0% | — |
| Chess Puzzles | 30% | — |
| EnigmaEval | 8.2% | — |
| Mystery Game Puzzles | 10% | — |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math GPT-5 Mini leads
GPT-5 Mini: 46.7 (#69), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | GPT-5 Mini | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 8.1% |
| Omni-MATH | 72.2% | 33% |
| LMArena Math | 1378 | 1283 |
| MATH Level 5 | 97.8% | 63.2% |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 9% | — |
| FrontierMath (Feb 2025 set) | 27.2% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5 Mini leads
GPT-5 Mini: 45.6 (#86), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | GPT-5 Mini | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 75% | 49.1% |
| MMLU-Pro | 83.5% | 63.1% |
| Confabulations | 13.3% | 19.1% |
| GPQA (HELM) | 75.6% | 42.6% |
| LMArena Expert | 1379 | 1245 |
| Humanity's Last Exam | 19.4% | — |
| SimpleQA Verified | 21.6% | — |
| Vectara Hallucination Rate | 12.9% | — |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multimodal Not comparable
GPT-5 Mini: 35.6 (#85), Qwen2.5 72B Instruct: —
| Benchmark | GPT-5 Mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1202 | — |
| VPCT | 40.2% | — |
Multilingual GPT-5 Mini leads
GPT-5 Mini: 48.9 (#137), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | GPT-5 Mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1363 | 1252 |
| LMArena Chinese | 1385 | 1272 |
| LMArena French | 1386 | 1280 |
| LMArena German | 1366 | 1234 |
| LMArena Japanese | 1341 | 1180 |
| LMArena Korean | 1308 | 1188 |
| LMArena Russian | 1362 | 1264 |
| LMArena Spanish | 1355 | 1256 |
Instruction Following GPT-5 Mini leads
GPT-5 Mini: 76.2 (#46), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | GPT-5 Mini | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 92.7% | 80.6% |
| LMArena Instruction Following | 1357 | 1254 |
Long Context GPT-5 Mini leads
GPT-5 Mini: 41.9 (#132), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | GPT-5 Mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1355 | 1282 |
| Fiction.LiveBench | 69.4% | — |
Writing & Preference GPT-5 Mini leads
GPT-5 Mini: 55.2 (#148), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | GPT-5 Mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1373 | 1269 |
| LMArena Creative Writing | 1325 | 1221 |
| WildBench | 85.5% | 80.2% |
| LMArena Multi-Turn | 1363 | 1272 |
| Short-Story Creative Writing | 83.1% | — |
| EQ-Bench Creative Writing | 1313 | — |
Frequently asked questions
Is GPT-5 Mini better than Qwen2.5 72B Instruct?
GPT-5 Mini is the stronger model overall, scoring 41.8 to 31.9 on the Noometry Index.
Which is cheaper, GPT-5 Mini or Qwen2.5 72B Instruct?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is GPT-5 Mini or Qwen2.5 72B Instruct better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.1 versus 33.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 131K.
How many benchmarks do GPT-5 Mini and Qwen2.5 72B Instruct share?
31 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Qwen2.5 72B Instruct has 43.