Model comparison
Grok 4 vs Qwen2.5 72B Instruct
Grok 4 is the stronger model overall, scoring 48.1 to 31.9 on the Noometry Index.
Last verified . 30 shared benchmarks.
Alibaba (Qwen)Summary
- They share 30 benchmarks with published results for both. Grok 4 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 Grok 4 leads 48.4 to 19.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 84% for Grok 4 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 4 | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 48.1 | 31.9 |
| Released | 2025-07-09 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 48 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4 leads
Grok 4: 50.3 (#46), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Grok 4 | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 45.7% | 16% |
| LMArena Coding | 1408 | 1292 |
| Aider Polyglot | 79.6% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Grok 4 leads
Grok 4: 32.3 (#68), Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Grok 4 | Qwen2.5 72B Instruct |
|---|---|---|
| BALROG | 43.6% | 16.2% |
| METR Time Horizons | 66.6% | 35.8% |
| Terminal-Bench | 27.2% | — |
| Berkeley Function Calling Leaderboard | 63% | — |
| GDPval | 21.1% | — |
| TheAgentCompany | — | 5.7% |
| Cybench | 43% | — |
| DeepResearch Bench | 47.3% | — |
| LMArena Search | 1142 | — |
Reasoning Grok 4 leads
Grok 4: 36.7 (#65), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Grok 4 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1409 | 1271 |
| Epoch Capabilities Index | 146.44 | 129 |
| ForecastBench | 60.9 | 57.5 |
| ARC-AGI-2 | 16% | — |
| SimpleBench | 60.5% | — |
| Kagi LLM Benchmark | 73.6% | — |
| ARC-AGI-1 | 66.7% | — |
| Chess Puzzles | 28% | — |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Grok 4 leads
Grok 4: 48.4 (#64), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Grok 4 | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 84% | 8.1% |
| Omni-MATH | 60.3% | 33% |
| LMArena Math | 1422 | 1283 |
| MATH Level 5 | — | 63.2% |
| FrontierMath (Feb 2025 set) | 19.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Grok 4 leads
Grok 4: 53.8 (#55), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Grok 4 | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 87% | 49.1% |
| MMLU-Pro | 85.1% | 63.1% |
| Confabulations | 12.4% | 19.1% |
| GPQA (HELM) | 72.7% | 42.6% |
| LMArena Expert | 1415 | 1245 |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multimodal Not comparable
Grok 4: 33.7 (#94), Qwen2.5 72B Instruct: —
| Benchmark | Grok 4 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1210 | — |
| GeoBench | 45% | — |
Multilingual Grok 4 leads
Grok 4: 51.8 (#103), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Grok 4 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1403 | 1252 |
| LMArena Chinese | 1427 | 1272 |
| LMArena French | 1418 | 1280 |
| LMArena German | 1429 | 1234 |
| LMArena Japanese | 1394 | 1180 |
| LMArena Korean | 1377 | 1188 |
| LMArena Russian | 1410 | 1264 |
| LMArena Spanish | 1420 | 1256 |
Instruction Following Grok 4 leads
Grok 4: 79.2 (#5), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Grok 4 | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 94.9% | 80.6% |
| LMArena Instruction Following | 1387 | 1254 |
Long Context Grok 4 leads
Grok 4: 63.1 (#4), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Grok 4 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1409 | 1282 |
| Fiction.LiveBench | 94.4% | — |
Writing & Preference Grok 4 leads
Grok 4: 58.5 (#116), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Grok 4 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1411 | 1269 |
| LMArena Creative Writing | 1397 | 1221 |
| WildBench | 79.7% | 80.2% |
| LMArena Multi-Turn | 1416 | 1272 |
| Short-Story Creative Writing | 76.9% | — |
Frequently asked questions
Is Grok 4 better than Qwen2.5 72B Instruct?
Grok 4 is the stronger model overall, scoring 48.1 to 31.9 on the Noometry Index.
Is Grok 4 or Qwen2.5 72B Instruct better for coding?
Grok 4 scores higher on coding benchmarks: 50.3 versus 33.2 in the Noometry coding category.
How many benchmarks do Grok 4 and Qwen2.5 72B Instruct share?
30 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Qwen2.5 72B Instruct has 43.