Model comparison
Grok 4.7 vs Qwen3 235B-A22B
Grok 4.7 is the stronger model overall, scoring 46.6 to 43.2 on the Noometry Index. Qwen3 235B-A22B costs 2.4× less per token, which makes it the better buy when Grok 4.7's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
Alibaba (Qwen)Summary
- They share 15 benchmarks with published results for both. Grok 4.7 scores higher in 4 categories and Qwen3 235B-A22B in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.7 leads 40.1 to 15.8.
- The biggest single-benchmark swing is CritPt: 17.7% for Grok 4.7 and 0% for Qwen3 235B-A22B.
- Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $2 / $6 for Grok 4.7.
- Grok 4.7 accepts more context: 500K tokens versus 131K.
- Qwen3 235B-A22B has downloadable open weights; the other is API-only.
Side by side
| Grok 4.7 | Qwen3 235B-A22B | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 46.6 | 43.2 |
| Released | 2026-09-21 | 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 | 21 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.7 leads
Grok 4.7: 55.0 (#23), Qwen3 235B-A22B: 43.8 (#72)
| Benchmark | Grok 4.7 | Qwen3 235B-A22B |
|---|---|---|
| SciCode | 57.4% | 42.4% |
| LMArena Coding | 1423 | 1445 |
| Aider Polyglot | — | 59.6% |
| CursorBench | 46.3% | — |
| LMArena WebDev | 1636 | — |
| FrontierSWE | 29.5% | — |
| WeirdML | — | 41% |
Agentic & Tool Use Not comparable
Grok 4.7: —, Qwen3 235B-A22B: 33.7 (#48)
| Benchmark | Grok 4.7 | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 52.1% |
| Vending-Bench 2 | — | -11.34 |
Reasoning Grok 4.7 leads
Grok 4.7: 40.1 (#49), Qwen3 235B-A22B: 15.8 (#302)
| Benchmark | Grok 4.7 | Qwen3 235B-A22B |
|---|---|---|
| CritPt | 17.7% | 0% |
| LMArena Hard Prompts | 1409 | 1433 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 31% |
| Kagi LLM Benchmark | — | 69.4% |
| NYT Connections (extended) | 76.8% | — |
| ARC-AGI-1 | — | 11% |
| Chess Puzzles | — | 12% |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LMCA | — | 29.3% |
| Epoch Capabilities Index | — | 143.87 |
| ForecastBench | — | 59.7 |
Math Qwen3 235B-A22B leads
Grok 4.7: 41.1 (#73), Qwen3 235B-A22B: 48.5 (#51)
| Benchmark | Grok 4.7 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Math | 1393 | 1431 |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| ProofBench | 34% | — |
| Omni-MATH | — | 71.8% |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3 235B-A22B leads
Grok 4.7: 38.7 (#130), Qwen3 235B-A22B: 49.5 (#67)
| Benchmark | Grok 4.7 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Expert | 1413 | 1460 |
| GPQA Diamond | — | 80.1% |
| SimpleQA Verified | — | 40.4% |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 72.7% |
Multimodal Not comparable
Grok 4.7: 28.8 (#107), Qwen3 235B-A22B: —
| Benchmark | Grok 4.7 | Qwen3 235B-A22B |
|---|---|---|
| Furniture Assembly | 20.8% | — |
Multilingual Qwen3 235B-A22B leads
Grok 4.7: 51.2 (#108), Qwen3 235B-A22B: 52.5 (#84)
| Benchmark | Grok 4.7 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1392 | 1410 |
| LMArena Chinese | 1476 | 1481 |
| LMArena Russian | 1396 | 1411 |
| LMArena French | — | 1445 |
| LMArena German | — | 1432 |
| LMArena Japanese | — | 1400 |
| LMArena Korean | — | 1391 |
| LMArena Spanish | — | 1429 |
Instruction Following Grok 4.7 leads
Grok 4.7: 74.3 (#98), Qwen3 235B-A22B: 72.8 (#132)
| Benchmark | Grok 4.7 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1406 | 1408 |
| IFEval | — | 83.5% |
Long Context Qwen3 235B-A22B leads
Grok 4.7: 43.2 (#97), Qwen3 235B-A22B: 46.1 (#25)
| Benchmark | Grok 4.7 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1416 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Grok 4.7 leads
Grok 4.7: 70.1 (#23), Qwen3 235B-A22B: 59.8 (#103)
| Benchmark | Grok 4.7 | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1397 | 1419 |
| LMArena Creative Writing | 1391 | 1384 |
| EQ-Bench Creative Writing | 2007 | 1366 |
| LMArena Multi-Turn | 1389 | 1432 |
| Short-Story Creative Writing | — | 83% |
| WildBench | — | 86.6% |
Frequently asked questions
Is Grok 4.7 better than Qwen3 235B-A22B?
Grok 4.7 is the stronger model overall, scoring 46.6 to 43.2 on the Noometry Index. Qwen3 235B-A22B costs 2.4× less per token, which makes it the better buy when Grok 4.7's lead doesn't matter for your workload.
Which is cheaper, Grok 4.7 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 4.7 lists at $2 and $6.
Is Grok 4.7 or Qwen3 235B-A22B better for coding?
Grok 4.7 scores higher on coding benchmarks: 55.0 versus 43.8 in the Noometry coding category.
Which has the bigger context window?
Grok 4.7 does, with 500K tokens against 131K.
How many benchmarks do Grok 4.7 and Qwen3 235B-A22B share?
15 benchmarks have published results for both models. Grok 4.7 has 21 scored results on Noometry and Qwen3 235B-A22B has 49.