Model comparison
Grok 4.20 (Non-Reasoning) vs Qwen2.5-VL 72B Instruct
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 29.9 on the Noometry Index.
Last verified . 2 shared benchmarks.
Alibaba (Qwen)Summary
- They share 2 benchmarks with published results for both. Grok 4.20 (Non-Reasoning) scores higher in 2 categories and Qwen2.5-VL 72B Instruct in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.3 to 20.7.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 75% for Grok 4.20 (Non-Reasoning) and 36% for Qwen2.5-VL 72B Instruct.
- Grok 4.20 (Non-Reasoning) is cheaper at $1.25 / $2.50 per million input/output tokens, against $2.80 / $8.40 for Qwen2.5-VL 72B Instruct.
- Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 131K.
- Qwen2.5-VL 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 48.6 | 29.9 |
| Released | 2026-02-17 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | 1M | 131K |
| Max output | 30K | 8K |
| Input $ / M tokens | $1.25 | $2.80 |
| Output $ / M tokens | $2.50 | $8.40 |
| Results tracked | 46 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Grok 4.20 (Non-Reasoning): 42.1 (#112), Qwen2.5-VL 72B Instruct: —
| Benchmark | Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena WebDev | 1375 | — |
| WeirdML | 52.3% | — |
| LMArena Coding | 1459 | — |
| ALE-Bench | 1,150 | — |
Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 34.4 (#46), Qwen2.5-VL 72B Instruct: 18.6 (#144)
| Benchmark | Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct |
|---|---|---|
| Terminal-Bench | 57.3% | — |
| τ²-bench Banking | 18% | — |
| OSWorld | — | 5% |
| LMArena Search | 1189 | — |
| Vending-Bench 2 | 4,663 | — |
Reasoning Grok 4.20 (Non-Reasoning) leads
Grok 4.20 (Non-Reasoning): 52.3 (#32), Qwen2.5-VL 72B Instruct: 20.7 (#233)
| Benchmark | Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct |
|---|---|---|
| Kagi LLM Benchmark | 75% | 36% |
| ARC-AGI-2 | 65.1% | — |
| NYT Connections (extended) | 85.4% | — |
| ARC-AGI-1 | 89.5% | — |
| Chess Puzzles | 24% | — |
| Thematic Generalization | 63.8% | — |
| LMArena Hard Prompts | 1451 | — |
| DTBench | 90.1% | — |
| LMCA | 38.7% | — |
| Epoch Capabilities Index | 151.98 | — |
| ForecastBench | 61.4 | — |
Math Not comparable
Grok 4.20 (Non-Reasoning): 48.2 (#65), Qwen2.5-VL 72B Instruct: —
| Benchmark | Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct |
|---|---|---|
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 17.1% | — |
| OTIS Mock AIME 2024-2025 | 92.2% | — |
| ProofBench | 14% | — |
| LMArena Math | 1455 | — |
Knowledge Not comparable
Grok 4.20 (Non-Reasoning): 52.8 (#60), Qwen2.5-VL 72B Instruct: —
| Benchmark | Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct |
|---|---|---|
| GPQA Diamond | 89.3% | — |
| SimpleQA Verified | 30.2% | — |
| LMArena Expert | 1439 | — |
Multimodal Too close to call
Grok 4.20 (Non-Reasoning): 33.3 (#98), Qwen2.5-VL 72B Instruct: 33.5 (#97)
| Benchmark | Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Vision | 1263 | 1107 |
| Video-MME | — | 73.5% |
| GeoBench | — | 62% |
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1416 | — |
| SpatialViz-Bench | — | 33.3% |
Multilingual Not comparable
Grok 4.20 (Non-Reasoning): 54.5 (#40), Qwen2.5-VL 72B Instruct: —
| Benchmark | Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Non-English | 1441 | — |
| LMArena Chinese | 1481 | — |
| LMArena French | 1476 | — |
| LMArena German | 1465 | — |
| LMArena Japanese | 1449 | — |
| LMArena Korean | 1417 | — |
| LMArena Russian | 1458 | — |
| LMArena Spanish | 1443 | — |
Instruction Following Not comparable
Grok 4.20 (Non-Reasoning): 74.8 (#83), Qwen2.5-VL 72B Instruct: —
| Benchmark | Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1420 | — |
Long Context Not comparable
Grok 4.20 (Non-Reasoning): 45.5 (#34), Qwen2.5-VL 72B Instruct: —
| Benchmark | Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct |
|---|---|---|
| CL-bench | 22.2% | — |
| CL-bench Life | 11.9% | — |
| LMArena Longer Query | 1437 | — |
Writing & Preference Not comparable
Grok 4.20 (Non-Reasoning): 65.7 (#44), Qwen2.5-VL 72B Instruct: —
| Benchmark | Grok 4.20 (Non-Reasoning) | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Text | 1451 | — |
| LMArena Creative Writing | 1438 | — |
| EQ-Bench Creative Writing | 1574 | — |
| LMArena Multi-Turn | 1456 | — |
Frequently asked questions
Is Grok 4.20 (Non-Reasoning) better than Qwen2.5-VL 72B Instruct?
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 29.9 on the Noometry Index.
Which is cheaper, Grok 4.20 (Non-Reasoning) or Qwen2.5-VL 72B Instruct?
Grok 4.20 (Non-Reasoning) is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; Qwen2.5-VL 72B Instruct lists at $2.80 and $8.40.
Which has the bigger context window?
Grok 4.20 (Non-Reasoning) does, with 1M tokens against 131K.
How many benchmarks do Grok 4.20 (Non-Reasoning) and Qwen2.5-VL 72B Instruct share?
2 benchmarks have published results for both models. Grok 4.20 (Non-Reasoning) has 46 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.