Model comparison
o4-mini vs Qwen2.5 72B Instruct
o4-mini is the stronger model overall, scoring 41.6 to 31.9 on the Noometry Index.
Last verified . 32 shared benchmarks.
Alibaba (Qwen)Summary
- They share 32 benchmarks with published results for both. o4-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 o4-mini leads 40.8 to 19.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 81.7% for o4-mini and 8.1% for Qwen2.5 72B Instruct.
- o4-mini is cheaper at $1.10 / $4.40 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- o4-mini accepts more context: 200K tokens versus 131K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| o4-mini | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 41.6 | 31.9 |
| Released | 2025-04-16 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | 200K | 131K |
| Max output | 100K | 8K |
| Input $ / M tokens | $1.10 | $1.40 |
| Output $ / M tokens | $4.40 | $5.60 |
| Results tracked | 60 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o4-mini leads
o4-mini: 40.9 (#127), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | o4-mini | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 52.6% | 16% |
| LMArena Coding | 1368 | 1292 |
| SWE-bench Verified (bash only) | 45% | — |
| Aider Polyglot | 72% | — |
| GSO | 3.6% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
| CadEval | 62% | — |
| ALE-Bench | 826.17 | — |
| AlgoTune | 1.72 | — |
Agentic & Tool Use o4-mini leads
o4-mini: 32.6 (#61), Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | o4-mini | Qwen2.5 72B Instruct |
|---|---|---|
| METR Time Horizons | 63.9% | 35.8% |
| Berkeley Function Calling Leaderboard | 53.2% | — |
| GDPval | 25.3% | — |
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
Reasoning o4-mini leads
o4-mini: 24.6 (#162), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | o4-mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1351 | 1271 |
| DTBench | 77.6% | 62.9% |
| LMCA | 26.5% | 13.4% |
| Epoch Capabilities Index | 145.64 | 129 |
| ForecastBench | 61.8 | 57.5 |
| ARC-AGI-2 | 6.1% | — |
| SimpleBench | 38.7% | — |
| Kagi LLM Benchmark | 67.6% | — |
| ARC-AGI-1 | 58.7% | — |
| CritPt | 0.6% | — |
| Chess Puzzles | 26% | — |
| EnigmaEval | 9.2% | — |
| Mystery Game Puzzles | 5% | — |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math o4-mini leads
o4-mini: 40.8 (#89), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | o4-mini | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 81.7% | 8.1% |
| Omni-MATH | 72% | 33% |
| LMArena Math | 1389 | 1283 |
| MATH Level 5 | 97.8% | 63.2% |
| FrontierMath (Tiers 1-3) | 36.1% | — |
| FrontierMath Tier 4 | 4.9% | — |
| FrontierMath (Feb 2025 set) | 24.8% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge o4-mini leads
o4-mini: 43.6 (#91), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | o4-mini | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 79.6% | 49.1% |
| MMLU-Pro | 82% | 63.1% |
| Confabulations | 15.8% | 19.1% |
| GPQA (HELM) | 73.5% | 42.6% |
| LMArena Expert | 1343 | 1245 |
| Humanity's Last Exam | 18.1% | — |
| SimpleQA Verified | 19.6% | — |
| Vectara Hallucination Rate | 18.6% | — |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multimodal Not comparable
o4-mini: 40.2 (#49), Qwen2.5 72B Instruct: —
| Benchmark | o4-mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1194 | — |
| GeoBench | 64% | — |
| VPCT | 57.5% | — |
Multilingual o4-mini leads
o4-mini: 47.0 (#154), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | o4-mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1337 | 1252 |
| LMArena Chinese | 1354 | 1272 |
| LMArena French | 1364 | 1280 |
| LMArena German | 1336 | 1234 |
| LMArena Japanese | 1308 | 1180 |
| LMArena Korean | 1312 | 1188 |
| LMArena Russian | 1334 | 1264 |
| LMArena Spanish | 1347 | 1256 |
Instruction Following o4-mini leads
o4-mini: 75.2 (#68), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | o4-mini | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 92.8% | 80.6% |
| LMArena Instruction Following | 1321 | 1254 |
Long Context o4-mini leads
o4-mini: 45.5 (#33), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | o4-mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1315 | 1282 |
| Fiction.LiveBench | 77.8% | — |
Writing & Preference o4-mini leads
o4-mini: 54.0 (#152), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | o4-mini | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1353 | 1269 |
| LMArena Creative Writing | 1294 | 1221 |
| WildBench | 85.4% | 80.2% |
| LMArena Multi-Turn | 1350 | 1272 |
| Short-Story Creative Writing | 75% | — |
Frequently asked questions
Is o4-mini better than Qwen2.5 72B Instruct?
o4-mini is the stronger model overall, scoring 41.6 to 31.9 on the Noometry Index.
Which is cheaper, o4-mini or Qwen2.5 72B Instruct?
o4-mini is cheaper. It lists at $1.10 per million input tokens and $4.40 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is o4-mini or Qwen2.5 72B Instruct better for coding?
o4-mini scores higher on coding benchmarks: 40.9 versus 33.2 in the Noometry coding category.
Which has the bigger context window?
o4-mini does, with 200K tokens against 131K.
How many benchmarks do o4-mini and Qwen2.5 72B Instruct share?
32 benchmarks have published results for both models. o4-mini has 60 scored results on Noometry and Qwen2.5 72B Instruct has 43.