Model comparison
Grok 3 vs Qwen2.5 72B Instruct
Grok 3 is the stronger model overall, scoring 39.9 to 31.9 on the Noometry Index.
Last verified . 29 shared benchmarks.
Alibaba (Qwen)Summary
- They share 29 benchmarks with published results for both. Grok 3 scores higher in 7 categories and Qwen2.5 72B Instruct in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok 3 leads 46.2 to 27.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 55.6% for Grok 3 and 8.1% for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| Grok 3 | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 39.9 | 31.9 |
| Released | 2025-04-09 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 40 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
Grok 3: 41.9 (#115), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Grok 3 | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 37.2% | 16% |
| LMArena Coding | 1432 | 1292 |
| Aider Polyglot | 53.3% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Grok 3 leads
Grok 3: 30.5 (#76), Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Grok 3 | Qwen2.5 72B Instruct |
|---|---|---|
| BALROG | 29.5% | 16.2% |
| TheAgentCompany | — | 5.7% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
Grok 3: 13.7 (#333), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Grok 3 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1434 | 1271 |
| Epoch Capabilities Index | 138.33 | 129 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 36.1% | — |
| Kagi LLM Benchmark | 61.3% | — |
| ARC-AGI-1 | 5.5% | — |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Grok 3 leads
Grok 3: 38.0 (#145), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Grok 3 | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 55.6% | 8.1% |
| Omni-MATH | 46.4% | 33% |
| LMArena Math | 1391 | 1283 |
| MATH Level 5 | 88.7% | 63.2% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Grok 3 leads
Grok 3: 46.2 (#82), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Grok 3 | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 75.8% | 49.1% |
| MMLU-Pro | 78.8% | 63.1% |
| Confabulations | 14.2% | 19.1% |
| GPQA (HELM) | 65% | 42.6% |
| LMArena Expert | 1421 | 1245 |
| Vectara Hallucination Rate | 5.8% | — |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multilingual Grok 3 leads
Grok 3: 52.3 (#87), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Grok 3 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1410 | 1252 |
| LMArena Chinese | 1448 | 1272 |
| LMArena French | 1460 | 1280 |
| LMArena German | 1431 | 1234 |
| LMArena Japanese | 1387 | 1180 |
| LMArena Korean | 1373 | 1188 |
| LMArena Russian | 1416 | 1264 |
| LMArena Spanish | 1417 | 1256 |
Instruction Following Grok 3 leads
Grok 3: 75.0 (#73), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Grok 3 | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 88.4% | 80.6% |
| LMArena Instruction Following | 1409 | 1254 |
Long Context Too close to call
Grok 3: 38.7 (#192), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Grok 3 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1439 | 1282 |
| Fiction.LiveBench | 58.3% | — |
Writing & Preference Grok 3 leads
Grok 3: 55.8 (#141), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Grok 3 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1426 | 1269 |
| LMArena Creative Writing | 1414 | 1221 |
| WildBench | 84.9% | 80.2% |
| LMArena Multi-Turn | 1425 | 1272 |
| Short-Story Creative Writing | 76.4% | — |
| EQ-Bench Creative Writing | 1186 | — |
Frequently asked questions
Is Grok 3 better than Qwen2.5 72B Instruct?
Grok 3 is the stronger model overall, scoring 39.9 to 31.9 on the Noometry Index.
Is Grok 3 or Qwen2.5 72B Instruct better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 33.2 in the Noometry coding category.
How many benchmarks do Grok 3 and Qwen2.5 72B Instruct share?
29 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Qwen2.5 72B Instruct has 43.