Grok 4 vs Qwen3 7
Grok 4 
Qwen3 7 Verdict
- Grok 4 and Qwen3 7 (max) are level on quality (59.8 vs 60.1).
- Grok 4 is stronger in coding, human preference, long context, maths, multimodal.
- Qwen3 7 (max) is stronger in agents & tools, instruction following, knowledge, reasoning, composite.
- Grok 4 is 2.4× cheaper ($1.56 vs $3.75 per 1M blended).
| Metric | Grok 4 | Qwen3 7 (max) |
|---|---|---|
| BenchLeader Index | 59.8 | 60.1 |
| Agents & tools score | 53.4 | 54.1 |
| Coding score | 61.2 | 60.9 |
| Human preference score | 59.8 | 57.1 |
| Instruction following score | 67.7 | 74.6 |
| Knowledge score | 56.0 | 62.9 |
| Long context score | 77.7 | 63.5 |
| Maths score | 58.0 | 57.4 |
| Multimodal score | 54.5 | – |
| Reasoning score | 58.3 | 66.8 |
| Composite score | – | 53.5 |
| Blended price $/M | $1.56 | $3.75 |
| Output speed | – | 166 tok/s |
| Time to first answer | – | 17.0 s |
| Context window | 256k | 1M |
| GPQA Diamond | 87.0% | 90.9% |
| FrontierMath Tiers 1–3 | – | 64.6% |
| FrontierMath Tier 4 | – | 34.1% |
| OTIS Mock AIME | 84.0% | 95.6% |
| SWE-bench Verified (Epoch) | – | 77.3% |
| SimpleQA Verified | – | 55.8% |
| Terminal-Bench | 27.2% | – |
| SimpleBench | 60.5% | 70.4% |
| Fiction.LiveBench 120k | 96.9% | – |
| SciCode | – | 48.8% |
| Cybench | 43.0% | – |
| WeirdML | 45.7% | – |
| APEX-Agents | 15.2% | – |
| ProofBench | – | 26.0% |
| Epoch Capabilities Index | 146.4 | 153.7 |
| LMArena Text | 1411 | 1474 |
| LMArena Hard Prompts | 1420 | 1495 |
| LMArena Coding | 1435 | 1525 |
| LMArena WebDev | – | 1517 |
| LMArena Vision | 1210 | – |
| LMArena Agent | – | -3.1 |
| LiveBench | – | 73.1% |
| LiveBench Reasoning | – | 83.3% |
| LiveBench Coding | – | 74.2% |
| LiveBench Agentic Coding | – | 43.6% |
| LiveBench Mathematics | – | 85.3% |
| LiveBench Data Analysis | – | 71.8% |
| LiveBench Language | – | 79.7% |
| AA Intelligence Index | – | 29.9 |
| IFBench | – | 80.5% |
| AA-LCR | – | 79.0% |
| AA-Omniscience | – | 13.5 |
| Terminal-Bench Hard | – | 50.8% |
| GPQA Diamond (AA) | – | 92.3% |
| Humanity's Last Exam (AA) | – | 40.5% |
| SciCode (AA) | – | 49.5% |
| τ²-Bench Telecom (AA) | – | 94.7% |
| AIME (Vals) | 90.6% | – |
| LiveCodeBench | 83.3% | 87.1% |
| MMLU-Pro | 85.3% | 89.3% |
| LegalBench | 83.2% | 84.9% |
| CorpFin | 66.0% | 63.7% |
| TaxEval | 65.1% | 75.3% |
| MedQA | 92.5% | – |
| MGSM | 90.9% | – |
| Terminal-Bench 2.1 (Vals) | – | 61.0% |
| SWE-bench (Vals) | 57.8% | 68.8% |
| GPQA Diamond (Vals) | 88.1% | 90.2% |
| Vals Index | – | 44.8 |
| IMO 2025 | 21.4% | – |
| MathArena Apex | 2.1% | – |
| EQ-Bench 4 | – | 1110 |
| Kagi LLM Benchmark | 73.6% | – |
| IFEval (HELM) | 94.9% | – |
| Omni-MATH (HELM) | 60.3% | – |
| WildBench (HELM) | 79.7% | – |
| MMLU-Pro (HELM) | 85.1% | – |
| GPQA Diamond (HELM) | 72.6% | – |
| HELM Capabilities mean | 78.5% | – |
| Aider Polyglot | 79.6% | – |
| ARC-AGI-1 | 79.6% | – |
| ARC-AGI-2 | 29.4% | – |
| BFCL Overall | 63.0% | – |
Data as of 2026-09-13. Best configuration of each model; every score links to its source on the model pages.
Grok 4 vs Qwen3 7: questions
- Is Grok 4 better than Qwen3 7?
- Grok 4 and Qwen3 7 (max) are level on quality (59.8 vs 60.1). The BenchLeader Index combines every independent quality benchmark; Qwen3 7 (max) is ahead overall as of 2026-09-13, but check the category scores for your use.
- Is Grok 4 better than Qwen3 7 for coding?
- Grok 4 scores higher in coding (61 vs 61 on the category index, where 50 is average).
- Is Grok 4 better than Qwen3 7 for agentic tasks?
- Qwen3 7 scores higher in agentic tasks (54 vs 53 on the category index, where 50 is average).
- Which is cheaper, Grok 4 or Qwen3 7?
- Grok 4 is cheaper: $1.56 against $3.75 per million tokens, blended at three input tokens per output token.
- Which has the larger context window?
- Qwen3 7 accepts more context: 1M against 256k tokens.