Model comparison
Grok 4.5 vs Qwen3.6 Max Preview
Grok 4.5 is the stronger model overall, scoring 55.0 to 51.5 on the Noometry Index.
Last verified . 25 shared benchmarks.
Alibaba (Qwen)Summary
- They share 25 benchmarks with published results for both. Grok 4.5 scores higher in 8 categories and Qwen3.6 Max Preview in 0 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.5 leads 56.1 to 41.7.
- The biggest single-benchmark swing is Chess Puzzles: 36% for Grok 4.5 and 20% for Qwen3.6 Max Preview.
- Both cost about the same: $2 input and $6 output per million tokens.
- Grok 4.5 accepts more context: 500K tokens versus 262K.
Side by side
| Grok 4.5 | Qwen3.6 Max Preview | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 55.0 | 51.5 |
| Released | 2026-07-08 | 2026-04-20 |
| Weights | Proprietary | Proprietary |
| Context window | 500K | 262K |
| Max output | 500K | 66K |
| Input $ / M tokens | $2 | $1.30 |
| Output $ / M tokens | $6 | $7.80 |
| Results tracked | 52 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.5 leads
Grok 4.5: 52.2 (#35), Qwen3.6 Max Preview: 48.7 (#54)
| Benchmark | Grok 4.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena WebDev | 1553 | 1482 |
| LMArena Coding | 1474 | 1471 |
| SWE-bench Verified | — | 76.7% |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.4% | — |
| SciCode | 54.1% | — |
| WeirdML | 46.4% | — |
| ALE-Bench | 1,309 | — |
Agentic & Tool Use Not comparable
Grok 4.5: 44.4 (#17), Qwen3.6 Max Preview: —
| Benchmark | Grok 4.5 | Qwen3.6 Max Preview |
|---|---|---|
| Vending-Bench 2 | 3,887 | 4,254 |
| APEX-Agents | 56.2% | — |
| τ²-bench Banking | 47.9% | — |
| PostTrainBench | 23.4% | — |
| GBAEval | 65.4% | — |
| GDP.pdf | 14% | — |
| LMArena Search | 1213 | — |
Reasoning Grok 4.5 leads
Grok 4.5: 56.1 (#25), Qwen3.6 Max Preview: 41.7 (#53)
| Benchmark | Grok 4.5 | Qwen3.6 Max Preview |
|---|---|---|
| SimpleBench | 70% | 63% |
| NYT Connections (extended) | 79.9% | 74.1% |
| Chess Puzzles | 36% | 20% |
| LMArena Hard Prompts | 1462 | 1457 |
| DTBench | 96.5% | 87.2% |
| LMCA | 45.2% | 42.5% |
| Epoch Capabilities Index | 153.92 | 149.24 |
| ARC-AGI-2 | 52.6% | — |
| Kagi LLM Benchmark | 83.5% | — |
| ARC-AGI-1 | 87.2% | — |
| CritPt | 15.4% | — |
| Mystery Game Puzzles | — | 19% |
| Surface Evolver Bench | 74.4% | — |
Math Grok 4.5 leads
Grok 4.5: 60.9 (#35), Qwen3.6 Max Preview: 54.1 (#46)
| Benchmark | Grok 4.5 | Qwen3.6 Max Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 97.8% | 91.1% |
| LMArena Math | 1459 | 1465 |
| FrontierMath (Tiers 1-3) | 57.2% | — |
| FrontierMath Tier 4 | 24.4% | — |
| ProofBench | 31% | — |
| FrontierMath (Feb 2025 set) | — | 23.1% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Grok 4.5 leads
Grok 4.5: 62.3 (#24), Qwen3.6 Max Preview: 57.6 (#39)
| Benchmark | Grok 4.5 | Qwen3.6 Max Preview |
|---|---|---|
| GPQA Diamond | 93.4% | 87.4% |
| SimpleQA Verified | 48.3% | 52% |
| LMArena Expert | 1466 | 1478 |
Multimodal Not comparable
Grok 4.5: 37.6 (#72), Qwen3.6 Max Preview: —
| Benchmark | Grok 4.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Vision | 1288 | — |
| Blueprint-Bench 2 | 27.3% | — |
| Furniture Assembly | 22.5% | — |
| LMArena Document | 1452 | — |
Multilingual Too close to call
Grok 4.5: 54.4 (#42), Qwen3.6 Max Preview: 54.2 (#48)
| Benchmark | Grok 4.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Non-English | 1440 | 1437 |
| LMArena Chinese | 1496 | 1487 |
| LMArena French | 1456 | 1449 |
| LMArena Russian | 1448 | 1445 |
| LMArena Spanish | 1450 | 1454 |
| LMArena German | 1446 | — |
| LMArena Japanese | 1428 | — |
| LMArena Korean | 1404 | — |
Instruction Following Too close to call
Grok 4.5: 76.0 (#48), Qwen3.6 Max Preview: 75.7 (#55)
| Benchmark | Grok 4.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Instruction Following | 1446 | 1438 |
Long Context Too close to call
Grok 4.5: 44.8 (#56), Qwen3.6 Max Preview: 44.6 (#61)
| Benchmark | Grok 4.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Longer Query | 1463 | 1457 |
Writing & Preference Grok 4.5 leads
Grok 4.5: 65.8 (#42), Qwen3.6 Max Preview: 63.8 (#60)
| Benchmark | Grok 4.5 | Qwen3.6 Max Preview |
|---|---|---|
| LMArena Text | 1448 | 1447 |
| LMArena Creative Writing | 1442 | 1435 |
| LMArena Multi-Turn | 1456 | 1456 |
| EQ-Bench Creative Writing | 1579 | — |
Frequently asked questions
Is Grok 4.5 better than Qwen3.6 Max Preview?
Grok 4.5 is the stronger model overall, scoring 55.0 to 51.5 on the Noometry Index.
Which is cheaper, Grok 4.5 or Qwen3.6 Max Preview?
Qwen3.6 Max Preview is cheaper. It lists at $1.30 per million input tokens and $7.80 per million output tokens; Grok 4.5 lists at $2 and $6.
Is Grok 4.5 or Qwen3.6 Max Preview better for coding?
Grok 4.5 scores higher on coding benchmarks: 52.2 versus 48.7 in the Noometry coding category.
Which has the bigger context window?
Grok 4.5 does, with 500K tokens against 262K.
How many benchmarks do Grok 4.5 and Qwen3.6 Max Preview share?
25 benchmarks have published results for both models. Grok 4.5 has 52 scored results on Noometry and Qwen3.6 Max Preview has 29.