Model comparison
Grok 3 vs Qwen3-Coder 480B-A35B Instruct
Grok 3 is the stronger model overall, scoring 39.9 to 38.1 on the Noometry Index.
Last verified . 19 shared benchmarks.
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. Grok 3 scores higher in 7 categories and Qwen3-Coder 480B-A35B Instruct in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3-Coder 480B-A35B Instruct leads 25.5 to 13.7.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 61.3% for Grok 3 and 49.5% for Qwen3-Coder 480B-A35B Instruct.
- Qwen3-Coder 480B-A35B Instruct has downloadable open weights; the other is API-only.
Side by side
| Grok 3 | Qwen3-Coder 480B-A35B Instruct | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 39.9 | 38.1 |
| Released | 2025-04-09 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | — | 262K |
| Max output | — | 66K |
| Input $ / M tokens | — | $1.50 |
| Output $ / M tokens | — | $7.50 |
| Results tracked | 40 | 25 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
Grok 3: 41.9 (#115), Qwen3-Coder 480B-A35B Instruct: 35.5 (#223)
| Benchmark | Grok 3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| WeirdML | 37.2% | 41.2% |
| LMArena Coding | 1432 | 1412 |
| SWE-bench Verified (bash only) | — | 55.4% |
| Aider Polyglot | 53.3% | — |
| LMArena WebDev | — | 1275 |
| GSO | — | 4.9% |
| ALE-Bench | — | 461.45 |
| AlgoTune | — | 1.44 |
Agentic & Tool Use Grok 3 leads
Grok 3: 30.5 (#76), Qwen3-Coder 480B-A35B Instruct: 23.9 (#123)
| Benchmark | Grok 3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Terminal-Bench | — | 27.2% |
| BALROG | 29.5% | — |
Reasoning Qwen3-Coder 480B-A35B Instruct leads
Grok 3: 13.7 (#333), Qwen3-Coder 480B-A35B Instruct: 25.5 (#149)
| Benchmark | Grok 3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| Kagi LLM Benchmark | 61.3% | 49.5% |
| LMArena Hard Prompts | 1434 | 1372 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 36.1% | — |
| ARC-AGI-1 | 5.5% | — |
| Epoch Capabilities Index | 138.33 | — |
Math Too close to call
Grok 3: 38.0 (#145), Qwen3-Coder 480B-A35B Instruct: 37.6 (#150)
| Benchmark | Grok 3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Math | 1391 | 1365 |
| OTIS Mock AIME 2024-2025 | 55.6% | — |
| Omni-MATH | 46.4% | — |
| MATH Level 5 | 88.7% | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Grok 3 leads
Grok 3: 46.2 (#82), Qwen3-Coder 480B-A35B Instruct: 37.0 (#162)
| Benchmark | Grok 3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Expert | 1421 | 1338 |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 78.8% | — |
| Confabulations | 14.2% | — |
| Vectara Hallucination Rate | 5.8% | — |
| GPQA (HELM) | 65% | — |
Multilingual Grok 3 leads
Grok 3: 52.3 (#87), Qwen3-Coder 480B-A35B Instruct: 47.7 (#148)
| Benchmark | Grok 3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Non-English | 1410 | 1346 |
| LMArena Chinese | 1448 | 1357 |
| LMArena French | 1460 | 1398 |
| LMArena German | 1431 | 1325 |
| LMArena Japanese | 1387 | 1310 |
| LMArena Korean | 1373 | 1305 |
| LMArena Russian | 1416 | 1366 |
| LMArena Spanish | 1417 | 1360 |
Instruction Following Grok 3 leads
Grok 3: 75.0 (#73), Qwen3-Coder 480B-A35B Instruct: 71.6 (#147)
| Benchmark | Grok 3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Instruction Following | 1409 | 1355 |
| IFEval | 88.4% | — |
Long Context Qwen3-Coder 480B-A35B Instruct leads
Grok 3: 38.7 (#192), Qwen3-Coder 480B-A35B Instruct: 42.0 (#131)
| Benchmark | Grok 3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Longer Query | 1439 | 1378 |
| Fiction.LiveBench | 58.3% | — |
Writing & Preference Too close to call
Grok 3: 55.8 (#141), Qwen3-Coder 480B-A35B Instruct: 55.3 (#147)
| Benchmark | Grok 3 | Qwen3-Coder 480B-A35B Instruct |
|---|---|---|
| LMArena Text | 1426 | 1357 |
| LMArena Creative Writing | 1414 | 1333 |
| LMArena Multi-Turn | 1425 | 1365 |
| Short-Story Creative Writing | 76.4% | — |
| EQ-Bench Creative Writing | 1186 | — |
| WildBench | 84.9% | — |
Frequently asked questions
Is Grok 3 better than Qwen3-Coder 480B-A35B Instruct?
Grok 3 is the stronger model overall, scoring 39.9 to 38.1 on the Noometry Index.
Is Grok 3 or Qwen3-Coder 480B-A35B Instruct better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 35.5 in the Noometry coding category.
How many benchmarks do Grok 3 and Qwen3-Coder 480B-A35B Instruct share?
19 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Qwen3-Coder 480B-A35B Instruct has 25.