Model comparison
Grok 4.6 vs Qwen2.5-Coder-32B
Grok 4.6 is the stronger model overall, scoring 56.9 to 33.4 on the Noometry Index. Qwen2.5-Coder-32B costs 4.0× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Last verified . 13 shared benchmarks.
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Grok 4.6 scores higher in 8 categories and Qwen2.5-Coder-32B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 21.2.
- Qwen2.5-Coder-32B is cheaper at $0.66 / $1 per million input/output tokens, against $2 / $6 for Grok 4.6.
- Grok 4.6 accepts more context: 500K tokens versus 33K.
- Qwen2.5-Coder-32B has downloadable open weights; the other is API-only.
Side by side
| Grok 4.6 | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 56.9 | 33.4 |
| Released | 2026-08-12 | 2024-09-18 |
| Weights | Proprietary | Open |
| Context window | 500K | 33K |
| Max output | 500K | 29K |
| Input $ / M tokens | $2 | $0.66 |
| Output $ / M tokens | $6 | $1 |
| Results tracked | 49 | 31 |
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), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Grok 4.6 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1465 | 1276 |
| DeepSWE | 67.5% | — |
| FrontierCode | 48% | — |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| CursorBench | 41.4% | — |
| LMArena WebDev | 1617 | — |
| FrontierSWE | 25.3% | — |
| SciCode | 56.5% | — |
| WeirdML | 67.3% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| ALE-Bench | 1,508 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
Grok 4.6: 39.4 (#27), Qwen2.5-Coder-32B: —
| Benchmark | Grok 4.6 | Qwen2.5-Coder-32B |
|---|---|---|
| APEX-Agents | 65.3% | — |
| GDP.pdf | 17.2% | — |
| Vending-Bench 2 | 9,047 | — |
Reasoning Grok 4.6 leads
Grok 4.6: 61.4 (#20), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Grok 4.6 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1447 | 1251 |
| Epoch Capabilities Index | 156.44 | 119.49 |
| ARC-AGI-2 | 67.1% | — |
| SimpleBench | 75.9% | — |
| NYT Connections (extended) | 80% | — |
| ARC-AGI-1 | 87.5% | — |
| CritPt | 19.7% | — |
| Chess Puzzles | 40% | — |
| EBR-Bench | 30.5% | — |
| LiveBench Reasoning | — | 42.1% |
| Mystery Game Puzzles | 34% | — |
| DTBench | 97.3% | — |
| LiveBench Data Analysis | — | 49.9% |
| LMCA | 48.5% | — |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Grok 4.6 leads
Grok 4.6: 67.0 (#24), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Grok 4.6 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1423 | 1251 |
| FrontierMath (Tiers 1-3) | 66% | — |
| FrontierMath Tier 4 | 31.7% | — |
| OTIS Mock AIME 2024-2025 | 99.2% | — |
| ProofBench | 51% | — |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge Grok 4.6 leads
Grok 4.6: 63.3 (#20), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Grok 4.6 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1467 | 1221 |
| GPQA Diamond | 94% | — |
| SimpleQA Verified | 49.3% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multimodal Not comparable
Grok 4.6: 43.6 (#23), Qwen2.5-Coder-32B: —
| Benchmark | Grok 4.6 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Vision | 1263 | — |
| Blueprint-Bench 2 | 33.2% | — |
| Furniture Assembly | 40% | — |
| LMArena Document | 1452 | — |
Multilingual Grok 4.6 leads
Grok 4.6: 53.0 (#74), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Grok 4.6 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1420 | 1205 |
| LMArena Chinese | 1480 | 1222 |
| LMArena Russian | 1422 | 1228 |
| LMArena French | 1461 | — |
| LMArena German | 1431 | — |
| LMArena Japanese | 1376 | — |
| LMArena Korean | 1397 | — |
| LMArena Spanish | 1404 | — |
Instruction Following Grok 4.6 leads
Grok 4.6: 75.4 (#63), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Grok 4.6 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1431 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Grok 4.6 leads
Grok 4.6: 44.5 (#66), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Grok 4.6 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1454 | 1251 |
Writing & Preference Grok 4.6 leads
Grok 4.6: 62.3 (#80), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Grok 4.6 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1428 | 1230 |
| LMArena Creative Writing | 1428 | 1174 |
| LMArena Multi-Turn | 1425 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Grok 4.6 better than Qwen2.5-Coder-32B?
Grok 4.6 is the stronger model overall, scoring 56.9 to 33.4 on the Noometry Index. Qwen2.5-Coder-32B costs 4.0× 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 Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is cheaper. It lists at $0.66 per million input tokens and $1 per million output tokens; Grok 4.6 lists at $2 and $6.
Is Grok 4.6 or Qwen2.5-Coder-32B better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Grok 4.6 does, with 500K tokens against 33K.
How many benchmarks do Grok 4.6 and Qwen2.5-Coder-32B share?
13 benchmarks have published results for both models. Grok 4.6 has 49 scored results on Noometry and Qwen2.5-Coder-32B has 31.