Model comparison
Grok-2 (Dec 2024) vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.2 to 33.4 on the Noometry Index.
Last verified . 27 shared benchmarks.
Alibaba (Qwen)Summary
- They share 27 benchmarks with published results for both. Grok-2 (Dec 2024) scores higher in 1 category and Qwen3 235B-A22B in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3 235B-A22B leads 48.5 to 19.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 11.5% for Grok-2 (Dec 2024) and 86.7% for Qwen3 235B-A22B.
- Qwen3 235B-A22B has downloadable open weights; the other is API-only.
Side by side
| Grok-2 (Dec 2024) | Qwen3 235B-A22B | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 33.4 | 43.2 |
| Released | 2024-08-13 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $2.80 |
| Results tracked | 34 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Grok-2 (Dec 2024): 32.9 (#254), Qwen3 235B-A22B: 43.8 (#72)
| Benchmark | Grok-2 (Dec 2024) | Qwen3 235B-A22B |
|---|---|---|
| WeirdML | 22.2% | 41% |
| LMArena Coding | 1288 | 1445 |
| Aider Polyglot | — | 59.6% |
| SciCode | — | 42.4% |
| LiveBench Coding | 46.4% | — |
Agentic & Tool Use Not comparable
Grok-2 (Dec 2024): —, Qwen3 235B-A22B: 33.7 (#48)
| Benchmark | Grok-2 (Dec 2024) | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 52.1% |
| Vending-Bench 2 | — | -11.34 |
Reasoning Too close to call
Grok-2 (Dec 2024): 16.4 (#294), Qwen3 235B-A22B: 15.8 (#302)
| Benchmark | Grok-2 (Dec 2024) | Qwen3 235B-A22B |
|---|---|---|
| SimpleBench | 22.7% | 31% |
| LMArena Hard Prompts | 1272 | 1433 |
| DTBench | 65.2% | 80.3% |
| Epoch Capabilities Index | 130.48 | 143.87 |
| ARC-AGI-2 | — | 1.3% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| CritPt | — | 0% |
| Chess Puzzles | — | 12% |
| LiveBench Reasoning | 54.8% | — |
| Mystery Game Puzzles | — | 9% |
| LiveBench Data Analysis | 54.5% | — |
| LMCA | — | 29.3% |
| ForecastBench | — | 59.7 |
| LiveBench | 54.3% | — |
Math Qwen3 235B-A22B leads
Grok-2 (Dec 2024): 19.6 (#278), Qwen3 235B-A22B: 48.5 (#51)
| Benchmark | Grok-2 (Dec 2024) | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 11.5% | 86.7% |
| LMArena Math | 1283 | 1431 |
| MATH Level 5 | 63.5% | 68.9% |
| FrontierMath (Feb 2025 set) | 0.7% | 8.5% |
| Omni-MATH | — | 71.8% |
| LiveBench Math | 54.9% | — |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
Grok-2 (Dec 2024): 29.8 (#225), Qwen3 235B-A22B: 49.5 (#67)
| Benchmark | Grok-2 (Dec 2024) | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 53.8% | 80.1% |
| Confabulations | 20.1% | 15.6% |
| LMArena Expert | 1254 | 1460 |
| SimpleQA Verified | — | 40.4% |
| MMLU-Pro | — | 84.4% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 72.7% |
Multilingual Qwen3 235B-A22B leads
Grok-2 (Dec 2024): 43.3 (#183), Qwen3 235B-A22B: 52.5 (#84)
| Benchmark | Grok-2 (Dec 2024) | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1282 | 1410 |
| LMArena Chinese | 1289 | 1481 |
| LMArena French | 1317 | 1445 |
| LMArena German | 1287 | 1432 |
| LMArena Japanese | 1244 | 1400 |
| LMArena Korean | 1237 | 1391 |
| LMArena Russian | 1286 | 1411 |
| LMArena Spanish | 1281 | 1429 |
Instruction Following Qwen3 235B-A22B leads
Grok-2 (Dec 2024): 67.1 (#197), Qwen3 235B-A22B: 72.8 (#132)
| Benchmark | Grok-2 (Dec 2024) | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1270 | 1408 |
| LiveBench Instruction Following | 69.6% | — |
| IFEval | — | 83.5% |
Long Context Qwen3 235B-A22B leads
Grok-2 (Dec 2024): 38.7 (#186), Qwen3 235B-A22B: 46.1 (#25)
| Benchmark | Grok-2 (Dec 2024) | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1276 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Qwen3 235B-A22B leads
Grok-2 (Dec 2024): 48.8 (#192), Qwen3 235B-A22B: 59.8 (#103)
| Benchmark | Grok-2 (Dec 2024) | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1305 | 1419 |
| LMArena Creative Writing | 1284 | 1384 |
| Short-Story Creative Writing | 63.6% | 83% |
| LMArena Multi-Turn | 1290 | 1432 |
| EQ-Bench Creative Writing | — | 1366 |
| WildBench | — | 86.6% |
| LiveBench Language | 45.6% | — |
Frequently asked questions
Is Grok-2 (Dec 2024) better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.2 to 33.4 on the Noometry Index.
Is Grok-2 (Dec 2024) or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 43.8 versus 32.9 in the Noometry coding category.
How many benchmarks do Grok-2 (Dec 2024) and Qwen3 235B-A22B share?
27 benchmarks have published results for both models. Grok-2 (Dec 2024) has 34 scored results on Noometry and Qwen3 235B-A22B has 49.