Model comparison
Grok-3 mini vs Qwen2-72B
Grok-3 mini is the stronger model overall, scoring 41.2 to 30.0 on the Noometry Index.
Last verified . 21 shared benchmarks.
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. Grok-3 mini scores higher in 7 categories and Qwen2-72B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok-3 mini leads 46.4 to 21.2.
- The biggest single-benchmark swing is MATH Level 5: 90.9% for Grok-3 mini and 39.1% for Qwen2-72B.
- Qwen2-72B has downloadable open weights; the other is API-only.
Side by side
| Grok-3 mini | Qwen2-72B | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 41.2 | 30.0 |
| Released | 2025-04-09 | 2024-06-07 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok-3 mini leads
Grok-3 mini: 40.8 (#131), Qwen2-72B: 29.1 (#310)
| Benchmark | Grok-3 mini | Qwen2-72B |
|---|---|---|
| WeirdML | 42.6% | 11.3% |
| LMArena Coding | 1379 | 1196 |
| Aider Polyglot | 49.3% | — |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
Agentic & Tool Use Not comparable
Grok-3 mini: —, Qwen2-72B: 17.0 (#146)
| Benchmark | Grok-3 mini | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
Grok-3 mini: 13.6 (#334), Qwen2-72B: 23.2 (#181)
| Benchmark | Grok-3 mini | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1375 | 1191 |
| Epoch Capabilities Index | 140.35 | 125.28 |
| ARC-AGI-2 | 0.4% | — |
| Kagi LLM Benchmark | 61.3% | — |
| ARC-AGI-1 | 16.5% | — |
Math Grok-3 mini leads
Grok-3 mini: 42.1 (#85), Qwen2-72B: 30.2 (#236)
| Benchmark | Grok-3 mini | Qwen2-72B |
|---|---|---|
| LMArena Math | 1386 | 1235 |
| MATH Level 5 | 90.9% | 39.1% |
| OTIS Mock AIME 2024-2025 | 77.8% | — |
| Omni-MATH | 31.8% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
Knowledge Grok-3 mini leads
Grok-3 mini: 46.4 (#81), Qwen2-72B: 21.2 (#275)
| Benchmark | Grok-3 mini | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 76.3% | 40.8% |
| LMArena Expert | 1395 | 1171 |
| MMLU-Pro | 79.9% | — |
| Confabulations | 10.8% | — |
| GPQA (HELM) | 67.5% | — |
| MMLU | — | 82.4% |
Multilingual Grok-3 mini leads
Grok-3 mini: 48.1 (#145), Qwen2-72B: 35.9 (#244)
| Benchmark | Grok-3 mini | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1352 | 1176 |
| LMArena Chinese | 1387 | 1240 |
| LMArena French | 1357 | 1170 |
| LMArena German | 1349 | 1151 |
| LMArena Japanese | 1342 | 1111 |
| LMArena Korean | 1335 | 1083 |
| LMArena Russian | 1353 | 1169 |
| LMArena Spanish | 1381 | 1169 |
Instruction Following Grok-3 mini leads
Grok-3 mini: 78.5 (#9), Qwen2-72B: 61.7 (#241)
| Benchmark | Grok-3 mini | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1357 | 1181 |
| IFEval | 95.1% | — |
Long Context Grok-3 mini leads
Grok-3 mini: 41.0 (#147), Qwen2-72B: 36.1 (#235)
| Benchmark | Grok-3 mini | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1372 | 1192 |
| Fiction.LiveBench | 66.7% | — |
Writing & Preference Grok-3 mini leads
Grok-3 mini: 52.5 (#169), Qwen2-72B: 40.8 (#241)
| Benchmark | Grok-3 mini | Qwen2-72B |
|---|---|---|
| LMArena Text | 1370 | 1203 |
| LMArena Creative Writing | 1342 | 1181 |
| LMArena Multi-Turn | 1355 | 1196 |
| Short-Story Creative Writing | 73.5% | — |
| WildBench | 65.1% | — |
Frequently asked questions
Is Grok-3 mini better than Qwen2-72B?
Grok-3 mini is the stronger model overall, scoring 41.2 to 30.0 on the Noometry Index.
Is Grok-3 mini or Qwen2-72B better for coding?
Grok-3 mini scores higher on coding benchmarks: 40.8 versus 29.1 in the Noometry coding category.
How many benchmarks do Grok-3 mini and Qwen2-72B share?
21 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Qwen2-72B has 26.