Model comparison
Grok 4 vs Qwen3.5 397B-A17B
Grok 4 is the stronger model overall, scoring 47.8 to 45.7 on the Noometry Index.
Last verified . 23 shared benchmarks.
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Grok 4 scores higher in 6 categories and Qwen3.5 397B-A17B in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in long context, where Grok 4 leads 63.1 to 44.1.
- The biggest single-benchmark swing is Chess Puzzles: 28% for Grok 4 and 13% for Qwen3.5 397B-A17B.
- Qwen3.5 397B-A17B has downloadable open weights; the other is API-only.
Side by side
| Grok 4 | Qwen3.5 397B-A17B | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 47.8 | 45.7 |
| Released | 2025-07-09 | 2026-02-01 |
| Weights | Proprietary | Open |
| Context window | — | 262K |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.60 |
| Output $ / M tokens | — | $3.60 |
| Results tracked | 48 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4 leads
Grok 4: 49.9 (#44), Qwen3.5 397B-A17B: 41.7 (#108)
| Benchmark | Grok 4 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Coding | 1408 | 1464 |
| Aider Polyglot | 79.6% | — |
| LMArena WebDev | — | 1400 |
| WeirdML | 45.7% | — |
Agentic & Tool Use Qwen3.5 397B-A17B leads
Grok 4: 32.1 (#62), Qwen3.5 397B-A17B: 33.2 (#51)
| Benchmark | Grok 4 | Qwen3.5 397B-A17B |
|---|---|---|
| Terminal-Bench | 27.2% | — |
| APEX-Agents | — | 24.9% |
| Berkeley Function Calling Leaderboard | 63% | — |
| GDPval | 21.1% | — |
| τ²-bench Airline | — | 81.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 84.4% |
| τ²-bench Telecom | — | 97.8% |
| Cybench | 43% | — |
| DeepResearch Bench | 47.3% | — |
| BALROG | 43.6% | — |
| LMArena Search | 1142 | — |
| METR Time Horizons | 66.6% | — |
Reasoning Grok 4 leads
Grok 4: 36.9 (#59), Qwen3.5 397B-A17B: 33.6 (#62)
| Benchmark | Grok 4 | Qwen3.5 397B-A17B |
|---|---|---|
| Kagi LLM Benchmark | 73.6% | 73.7% |
| Chess Puzzles | 28% | 13% |
| LMArena Hard Prompts | 1409 | 1448 |
| Epoch Capabilities Index | 146.46 | 146.64 |
| ARC-AGI-2 | 16% | — |
| SimpleBench | 60.5% | — |
| NYT Connections (extended) | — | 58.9% |
| ARC-AGI-1 | 66.7% | — |
| Thematic Generalization | — | 65.1% |
| Mystery Game Puzzles | — | 18% |
| DTBench | — | 87.5% |
| LMCA | — | 37.9% |
| ForecastBench | 60.9 | — |
Math Grok 4 leads
Grok 4: 46.5 (#58), Qwen3.5 397B-A17B: 44.9 (#64)
| Benchmark | Grok 4 | Qwen3.5 397B-A17B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 84% | 88.9% |
| LMArena Math | 1422 | 1455 |
| FrontierMath (Tiers 1-3) | — | 31.2% |
| Omni-MATH | 60.3% | — |
| FrontierMath (Feb 2025 set) | 19.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Too close to call
Grok 4: 53.7 (#50), Qwen3.5 397B-A17B: 53.3 (#53)
| Benchmark | Grok 4 | Qwen3.5 397B-A17B |
|---|---|---|
| GPQA Diamond | 87% | 86.4% |
| LMArena Expert | 1417 | 1461 |
| MMLU-Pro | 85.1% | — |
| Confabulations | 12.4% | — |
| GPQA (HELM) | 72.7% | — |
Multimodal Qwen3.5 397B-A17B leads
Grok 4: 33.1 (#89), Qwen3.5 397B-A17B: 40.0 (#42)
| Benchmark | Grok 4 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | 1211 | 1262 |
| GeoBench | 45% | — |
Multilingual Qwen3.5 397B-A17B leads
Grok 4: 52.0 (#98), Qwen3.5 397B-A17B: 54.0 (#56)
| Benchmark | Grok 4 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | 1403 | 1431 |
| LMArena Chinese | 1428 | 1501 |
| LMArena French | 1418 | 1461 |
| LMArena German | 1429 | 1447 |
| LMArena Japanese | 1396 | 1427 |
| LMArena Korean | 1376 | 1382 |
| LMArena Russian | 1409 | 1429 |
| LMArena Spanish | 1419 | 1443 |
Instruction Following Grok 4 leads
Grok 4: 79.4 (#4), Qwen3.5 397B-A17B: 75.2 (#73)
| Benchmark | Grok 4 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | 1387 | 1424 |
| IFEval | 94.9% | — |
Long Context Grok 4 leads
Grok 4: 63.1 (#4), Qwen3.5 397B-A17B: 44.1 (#71)
| Benchmark | Grok 4 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | 1409 | 1442 |
| Fiction.LiveBench | 94.4% | — |
Writing & Preference Qwen3.5 397B-A17B leads
Grok 4: 58.7 (#111), Qwen3.5 397B-A17B: 62.5 (#74)
| Benchmark | Grok 4 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | 1412 | 1438 |
| LMArena Creative Writing | 1397 | 1401 |
| LMArena Multi-Turn | 1416 | 1447 |
| Short-Story Creative Writing | 76.9% | — |
| EQ-Bench Creative Writing | — | 1478 |
| WildBench | 79.7% | — |
Frequently asked questions
Is Grok 4 better than Qwen3.5 397B-A17B?
Grok 4 is the stronger model overall, scoring 47.8 to 45.7 on the Noometry Index.
Is Grok 4 or Qwen3.5 397B-A17B better for coding?
Grok 4 scores higher on coding benchmarks: 49.9 versus 41.7 in the Noometry coding category.
How many benchmarks do Grok 4 and Qwen3.5 397B-A17B share?
23 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Qwen3.5 397B-A17B has 36.