Model comparison
GPT-4.1 mini vs Qwen2.5 72B Instruct
GPT-4.1 mini is the stronger model overall, scoring 33.5 to 31.5 on the Noometry Index.
Last verified . 30 shared benchmarks.
Alibaba (Qwen)Summary
- They share 30 benchmarks with published results for both. GPT-4.1 mini scores higher in 6 categories and Qwen2.5 72B Instruct in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-4.1 mini leads 33.0 to 22.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 44.7% for GPT-4.1 mini and 8.1% for Qwen2.5 72B Instruct.
- GPT-4.1 mini is cheaper at $0.40 / $1.60 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- GPT-4.1 mini accepts more context: 1.05M tokens versus 131K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-4.1 mini | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 33.5 | 31.5 |
| Released | 2025-04-14 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 33K | 8K |
| Input $ / M tokens | $0.40 | $1.40 |
| Output $ / M tokens | $1.60 | $5.60 |
| Results tracked | 47 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 72B Instruct leads
GPT-4.1 mini: 30.4 (#286), Qwen2.5 72B Instruct: 32.8 (#256)
| Benchmark | GPT-4.1 mini | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 37.6% | 16% |
| BigCodeBench Instruct | 48.9% | 45.8% |
| LMArena Coding | 1367 | 1293 |
| SWE-bench Verified (bash only) | 23.9% | — |
| Aider Polyglot | 32.4% | — |
| SciCode | 40.4% | — |
| BigCodeBench Complete | — | 55.9% |
| CadEval | 16% | — |
Agentic & Tool Use GPT-4.1 mini leads
GPT-4.1 mini: 33.0 (#52), Qwen2.5 72B Instruct: 22.0 (#124)
| Benchmark | GPT-4.1 mini | Qwen2.5 72B Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 50.5% | — |
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
GPT-4.1 mini: 10.9 (#335), Qwen2.5 72B Instruct: 21.4 (#197)
| Benchmark | GPT-4.1 mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1349 | 1271 |
| DTBench | 68.8% | 62.9% |
| LMCA | 21.1% | 13.4% |
| Epoch Capabilities Index | 135.01 | 129.01 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 48.6% | — |
| ARC-AGI-1 | 3.5% | — |
| CritPt | 0% | — |
| Chess Puzzles | 7% | — |
| Mystery Game Puzzles | 7% | — |
| BIG-Bench Hard | — | 79.8% |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math GPT-4.1 mini leads
GPT-4.1 mini: 23.1 (#264), Qwen2.5 72B Instruct: 18.1 (#281)
| Benchmark | GPT-4.1 mini | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 44.7% | 8.1% |
| Omni-MATH | 49.1% | 33% |
| LMArena Math | 1343 | 1283 |
| MATH Level 5 | 87.3% | 63.2% |
| FrontierMath (Tiers 1-3) | 6.7% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
Knowledge GPT-4.1 mini leads
GPT-4.1 mini: 34.7 (#183), Qwen2.5 72B Instruct: 26.9 (#247)
| Benchmark | GPT-4.1 mini | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 65.8% | 49.1% |
| MMLU-Pro | 78.3% | 63.1% |
| GPQA (HELM) | 61.4% | 42.6% |
| LMArena Expert | 1338 | 1245 |
| SimpleQA Verified | 12.7% | — |
| Confabulations | — | 19.1% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multimodal Not comparable
GPT-4.1 mini: 35.2 (#77), Qwen2.5 72B Instruct: —
| Benchmark | GPT-4.1 mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1181 | — |
Multilingual GPT-4.1 mini leads
GPT-4.1 mini: 45.9 (#161), Qwen2.5 72B Instruct: 41.2 (#208)
| Benchmark | GPT-4.1 mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1318 | 1252 |
| LMArena Chinese | 1329 | 1272 |
| LMArena French | 1357 | 1279 |
| LMArena German | 1349 | 1234 |
| LMArena Japanese | 1291 | 1180 |
| LMArena Korean | 1298 | 1188 |
| LMArena Russian | 1324 | 1264 |
| LMArena Spanish | 1319 | 1255 |
Instruction Following GPT-4.1 mini leads
GPT-4.1 mini: 73.9 (#113), Qwen2.5 72B Instruct: 65.7 (#216)
| Benchmark | GPT-4.1 mini | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 90.4% | 80.6% |
| LMArena Instruction Following | 1332 | 1254 |
Long Context Qwen2.5 72B Instruct leads
GPT-4.1 mini: 31.8 (#270), Qwen2.5 72B Instruct: 38.9 (#183)
| Benchmark | GPT-4.1 mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1344 | 1282 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference GPT-4.1 mini leads
GPT-4.1 mini: 48.8 (#193), Qwen2.5 72B Instruct: 46.9 (#210)
| Benchmark | GPT-4.1 mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1341 | 1269 |
| LMArena Creative Writing | 1300 | 1221 |
| WildBench | 83.8% | 80.2% |
| LMArena Multi-Turn | 1354 | 1272 |
| EQ-Bench Creative Writing | 1147 | — |
Frequently asked questions
Is GPT-4.1 mini better than Qwen2.5 72B Instruct?
GPT-4.1 mini is the stronger model overall, scoring 33.5 to 31.5 on the Noometry Index.
Which is cheaper, GPT-4.1 mini or Qwen2.5 72B Instruct?
GPT-4.1 mini is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is GPT-4.1 mini or Qwen2.5 72B Instruct better for coding?
Qwen2.5 72B Instruct scores higher on coding benchmarks: 32.8 versus 30.4 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 mini does, with 1.05M tokens against 131K.
How many benchmarks do GPT-4.1 mini and Qwen2.5 72B Instruct share?
30 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and Qwen2.5 72B Instruct has 43.