Model comparison
Grok 4.6 vs Qwen3 235B-A22B
Grok 4.6 is the stronger model overall, scoring 56.9 to 43.5 on the Noometry Index. Qwen3 235B-A22B costs 2.4× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
Alibaba (Qwen)Summary
- They share 32 benchmarks with published results for both. Grok 4.6 scores higher in 8 categories and Qwen3 235B-A22B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 15.7.
- The biggest single-benchmark swing is ARC-AGI-1: 87.5% for Grok 4.6 and 11% for Qwen3 235B-A22B.
- Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $2 / $6 for Grok 4.6.
- Grok 4.6 accepts more context: 500K tokens versus 131K.
- Qwen3 235B-A22B has downloadable open weights; the other is API-only.
Side by side
| Grok 4.6 | Qwen3 235B-A22B | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 56.9 | 43.5 |
| Released | 2026-08-12 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | 500K | 131K |
| Max output | 500K | 16K |
| Input $ / M tokens | $2 | $0.70 |
| Output $ / M tokens | $6 | $2.80 |
| Results tracked | 49 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.6 leads
Grok 4.6: 58.5 (#16), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | Grok 4.6 | Qwen3 235B-A22B |
|---|---|---|
| SciCode | 56.5% | 42.4% |
| WeirdML | 67.3% | 41% |
| LMArena Coding | 1465 | 1445 |
| DeepSWE | 67.5% | — |
| FrontierCode | 48% | — |
| Aider Polyglot | — | 59.6% |
| CursorBench | 41.4% | — |
| LMArena WebDev | 1617 | — |
| FrontierSWE | 25.3% | — |
| ALE-Bench | 1,508 | — |
Agentic & Tool Use Grok 4.6 leads
Grok 4.6: 39.4 (#27), Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | Grok 4.6 | Qwen3 235B-A22B |
|---|---|---|
| Vending-Bench 2 | 9,047 | -11.34 |
| APEX-Agents | 65.3% | — |
| Berkeley Function Calling Leaderboard | — | 52.1% |
| GDP.pdf | 17.2% | — |
Reasoning Grok 4.6 leads
Grok 4.6: 61.4 (#20), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | Grok 4.6 | Qwen3 235B-A22B |
|---|---|---|
| ARC-AGI-2 | 67.1% | 1.3% |
| SimpleBench | 75.9% | 31% |
| ARC-AGI-1 | 87.5% | 11% |
| CritPt | 19.7% | 0% |
| Chess Puzzles | 40% | 12% |
| LMArena Hard Prompts | 1447 | 1433 |
| Mystery Game Puzzles | 34% | 9% |
| DTBench | 97.3% | 80.3% |
| LMCA | 48.5% | 29.3% |
| Epoch Capabilities Index | 156.44 | 143.85 |
| Kagi LLM Benchmark | — | 69.4% |
| NYT Connections (extended) | 80% | — |
| EBR-Bench | 30.5% | — |
| ForecastBench | — | 59.7 |
Math Grok 4.6 leads
Grok 4.6: 67.0 (#24), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | Grok 4.6 | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 99.2% | 86.7% |
| LMArena Math | 1423 | 1432 |
| FrontierMath (Tiers 1-3) | 66% | — |
| FrontierMath Tier 4 | 31.7% | — |
| ProofBench | 51% | — |
| Omni-MATH | — | 71.8% |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Grok 4.6 leads
Grok 4.6: 63.3 (#20), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | Grok 4.6 | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 94% | 80.1% |
| SimpleQA Verified | 49.3% | 40.4% |
| LMArena Expert | 1467 | 1463 |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 72.7% |
Multimodal Not comparable
Grok 4.6: 43.6 (#23), Qwen3 235B-A22B: —
| Benchmark | Grok 4.6 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Vision | 1263 | — |
| Blueprint-Bench 2 | 33.2% | — |
| Furniture Assembly | 40% | — |
| LMArena Document | 1452 | — |
Multilingual Too close to call
Grok 4.6: 53.0 (#74), Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | Grok 4.6 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1420 | 1409 |
| LMArena Chinese | 1480 | 1481 |
| LMArena French | 1461 | 1445 |
| LMArena German | 1431 | 1433 |
| LMArena Japanese | 1376 | 1399 |
| LMArena Korean | 1397 | 1391 |
| LMArena Russian | 1422 | 1411 |
| LMArena Spanish | 1404 | 1430 |
Instruction Following Grok 4.6 leads
Grok 4.6: 75.4 (#63), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | Grok 4.6 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1431 | 1408 |
| IFEval | — | 83.5% |
Long Context Qwen3 235B-A22B leads
Grok 4.6: 44.5 (#66), Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | Grok 4.6 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1454 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Grok 4.6 leads
Grok 4.6: 62.3 (#80), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | Grok 4.6 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1428 | 1419 |
| LMArena Creative Writing | 1428 | 1384 |
| LMArena Multi-Turn | 1425 | 1432 |
| Short-Story Creative Writing | — | 83% |
| EQ-Bench Creative Writing | — | 1366 |
| WildBench | — | 86.6% |
Frequently asked questions
Is Grok 4.6 better than Qwen3 235B-A22B?
Grok 4.6 is the stronger model overall, scoring 56.9 to 43.5 on the Noometry Index. Qwen3 235B-A22B costs 2.4× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Which is cheaper, Grok 4.6 or Qwen3 235B-A22B?
Qwen3 235B-A22B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; Grok 4.6 lists at $2 and $6.
Is Grok 4.6 or Qwen3 235B-A22B better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 44.3 in the Noometry coding category.
Which has the bigger context window?
Grok 4.6 does, with 500K tokens against 131K.
How many benchmarks do Grok 4.6 and Qwen3 235B-A22B share?
32 benchmarks have published results for both models. Grok 4.6 has 49 scored results on Noometry and Qwen3 235B-A22B has 49.