Model comparison
Grok 4.7 vs Qwen3.5 35B-A3B
Grok 4.7 is the stronger model overall, scoring 53.1 to 42.0 on the Noometry Index. Qwen3.5 35B-A3B costs 4.4× less per token, which makes it the better buy when Grok 4.7's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Grok 4.7 scores higher in 8 categories and Qwen3.5 35B-A3B in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.7 leads 49.1 to 24.6.
- The biggest single-benchmark swing is SciCode: 57.8% for Grok 4.7 and 29.3% for Qwen3.5 35B-A3B.
- Qwen3.5 35B-A3B is cheaper at $0.25 / $2 per million input/output tokens, against $2 / $6 for Grok 4.7.
- Grok 4.7 accepts more context: 500K tokens versus 262K.
- Qwen3.5 35B-A3B has downloadable open weights; the other is API-only.
Side by side
| Grok 4.7 | Qwen3.5 35B-A3B | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 53.1 | 42.0 |
| Released | 2026-09-21 | 2026-02-01 |
| Weights | Proprietary | Open |
| Context window | 500K | 262K |
| Max output | 500K | 66K |
| Input $ / M tokens | $2 | $0.25 |
| Output $ / M tokens | $6 | $2 |
| Results tracked | 39 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.7 leads
Grok 4.7: 58.0 (#18), Qwen3.5 35B-A3B: 33.8 (#251)
| Benchmark | Grok 4.7 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena WebDev | 1639 | 1254 |
| SciCode | 57.8% | 29.3% |
| LMArena Coding | 1427 | 1410 |
| FrontierCode | 47.6% | — |
| CursorBench | 46.3% | — |
| FrontierSWE | 29.5% | — |
Agentic & Tool Use Not comparable
Grok 4.7: 36.7 (#37), Qwen3.5 35B-A3B: —
| Benchmark | Grok 4.7 | Qwen3.5 35B-A3B |
|---|---|---|
| APEX-Agents | 54.6% | — |
| GDP.pdf | 22.8% | — |
| Vending-Bench 2 | 10,537 | — |
Reasoning Grok 4.7 leads
Grok 4.7: 49.1 (#40), Qwen3.5 35B-A3B: 24.6 (#161)
| Benchmark | Grok 4.7 | Qwen3.5 35B-A3B |
|---|---|---|
| CritPt | 18% | 0.6% |
| Chess Puzzles | 38% | 10% |
| LMArena Hard Prompts | 1413 | 1400 |
| DTBench | 96% | 80% |
| LMCA | 49.4% | 29.5% |
| Epoch Capabilities Index | 153.53 | 142.52 |
| NYT Connections (extended) | 76.8% | — |
| Mystery Game Puzzles | 29% | — |
Math Grok 4.7 leads
Grok 4.7: 57.8 (#39), Qwen3.5 35B-A3B: 39.9 (#97)
| Benchmark | Grok 4.7 | Qwen3.5 35B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.1% | 70% |
| LMArena Math | 1407 | 1404 |
| FrontierMath (Tiers 1-3) | 53% | — |
| FrontierMath Tier 4 | 17.1% | — |
| MathArena Final-Answer Competitions | — | 56% |
| ProofBench | 34% | — |
Knowledge Grok 4.7 leads
Grok 4.7: 62.8 (#22), Qwen3.5 35B-A3B: 47.8 (#79)
| Benchmark | Grok 4.7 | Qwen3.5 35B-A3B |
|---|---|---|
| GPQA Diamond | 92.7% | 83.5% |
| LMArena Expert | 1422 | 1408 |
| SimpleQA Verified | 56% | — |
| Vectara Hallucination Rate | — | 10.5% |
Multimodal Not comparable
Grok 4.7: 35.5 (#87), Qwen3.5 35B-A3B: —
| Benchmark | Grok 4.7 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Vision | 1228 | — |
| Blueprint-Bench 2 | 32.5% | — |
| Furniture Assembly | 20.8% | — |
Multilingual Too close to call
Grok 4.7: 50.8 (#116), Qwen3.5 35B-A3B: 50.0 (#127)
| Benchmark | Grok 4.7 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Non-English | 1389 | 1378 |
| LMArena Chinese | 1455 | 1457 |
| LMArena French | 1455 | 1412 |
| LMArena Russian | 1397 | 1376 |
| LMArena Spanish | 1400 | 1392 |
| LMArena German | — | 1367 |
| LMArena Japanese | — | 1325 |
| LMArena Korean | — | 1356 |
Instruction Following Grok 4.7 leads
Grok 4.7: 74.1 (#105), Qwen3.5 35B-A3B: 72.8 (#128)
| Benchmark | Grok 4.7 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Instruction Following | 1404 | 1379 |
Long Context Too close to call
Grok 4.7: 43.1 (#104), Qwen3.5 35B-A3B: 42.4 (#127)
| Benchmark | Grok 4.7 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Longer Query | 1413 | 1389 |
Writing & Preference Grok 4.7 leads
Grok 4.7: 70.0 (#24), Qwen3.5 35B-A3B: 57.9 (#124)
| Benchmark | Grok 4.7 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Text | 1399 | 1395 |
| LMArena Creative Writing | 1391 | 1346 |
| LMArena Multi-Turn | 1393 | 1390 |
| EQ-Bench Creative Writing | 2007 | — |
Frequently asked questions
Is Grok 4.7 better than Qwen3.5 35B-A3B?
Grok 4.7 is the stronger model overall, scoring 53.1 to 42.0 on the Noometry Index. Qwen3.5 35B-A3B costs 4.4× 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 Qwen3.5 35B-A3B?
Qwen3.5 35B-A3B is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Grok 4.7 lists at $2 and $6.
Is Grok 4.7 or Qwen3.5 35B-A3B better for coding?
Grok 4.7 scores higher on coding benchmarks: 58.0 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
Grok 4.7 does, with 500K tokens against 262K.
How many benchmarks do Grok 4.7 and Qwen3.5 35B-A3B share?
23 benchmarks have published results for both models. Grok 4.7 has 39 scored results on Noometry and Qwen3.5 35B-A3B has 28.