# Grok 4.6 vs Qwen2.5 7B Instruct

> Grok 4.6 is the stronger model overall, scoring 56.9 to 29.0 on the Noometry Index. Qwen2.5 7B Instruct costs 9.8× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/grok-4-6-vs-qwen2-5-7b-instruct
- Last updated: 2026-10-12
- Shared benchmarks: 6

## Summary

- They share 6 benchmarks with published results for both. Grok 4.6 scores higher in 7 categories and Qwen2.5 7B Instruct in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4.6 leads 67.0 to 12.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 99.2% for Grok 4.6 and 2.5% for Qwen2.5 7B Instruct.
- Qwen2.5 7B Instruct is cheaper at $0.17 / $0.70 per million input/output tokens, against $2 / $6 for Grok 4.6.
- Grok 4.6 accepts more context: 500K tokens versus 131K.
- Qwen2.5 7B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 56.9 | 29.0 |
| Rank | 21 | 320 |
| Context | 500K | 131K |
| Input $/M | $2 | $0.17 |
| Output $/M | $6 | $0.70 |
| Weights | Proprietary | Open |

## Coding

- Grok 4.6: 58.5 (#16)
- Qwen2.5 7B Instruct: 36.5 (#208)

| Benchmark | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| DeepSWE | 67.5% | — |
| FrontierCode | 48% | — |
| CursorBench | 41.4% | — |
| LMArena WebDev | 1617 | — |
| FrontierSWE | 25.3% | — |
| SciCode | 56.5% | — |
| WeirdML | 67.3% | — |
| BigCodeBench Instruct | — | 37.6% |
| LMArena Coding | 1465 | — |
| BigCodeBench Complete | — | 46.1% |
| ALE-Bench | 1,508 | — |

## Agentic & Tool Use

- Grok 4.6: 39.4 (#27)
- Qwen2.5 7B Instruct: 23.8 (#124)

| Benchmark | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| APEX-Agents | 65.3% | — |
| BALROG | — | 7.8% |
| GDP.pdf | 17.2% | — |
| Vending-Bench 2 | 9,047 | — |

## Reasoning

- Grok 4.6: 61.4 (#20)
- Qwen2.5 7B Instruct: 14.8 (#322)

| Benchmark | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| Chess Puzzles | 40% | 0% |
| DTBench | 97.3% | 47.7% |
| LMCA | 48.5% | 6.4% |
| Epoch Capabilities Index | 156.44 | 118.51 |
| ARC-AGI-2 | 67.1% | — |
| SimpleBench | 75.9% | — |
| NYT Connections (extended) | 80% | — |
| ARC-AGI-1 | 87.5% | — |
| CritPt | 19.7% | — |
| EBR-Bench | 30.5% | — |
| LMArena Hard Prompts | 1447 | — |
| Mystery Game Puzzles | 34% | — |

## Math

- Grok 4.6: 67.0 (#24)
- Qwen2.5 7B Instruct: 12.6 (#306)

| Benchmark | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 99.2% | 2.5% |
| FrontierMath (Tiers 1-3) | 66% | — |
| FrontierMath Tier 4 | 31.7% | — |
| ProofBench | 51% | — |
| Omni-MATH | — | 29.4% |
| LMArena Math | 1423 | — |

## Knowledge

- Grok 4.6: 63.3 (#20)
- Qwen2.5 7B Instruct: 17.0 (#286)

| Benchmark | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| GPQA Diamond | 94% | 35.5% |
| SimpleQA Verified | 49.3% | — |
| MMLU-Pro | — | 53.9% |
| GPQA (HELM) | — | 34.1% |
| LMArena Expert | 1467 | — |
| MMLU | — | 72.9% |

## Multimodal

- Grok 4.6: 43.6 (#23)
- Qwen2.5 7B Instruct: —

| Benchmark | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Vision | 1263 | — |
| Blueprint-Bench 2 | 33.2% | — |
| Furniture Assembly | 40% | — |
| LMArena Document | 1452 | — |

## Multilingual

- Grok 4.6: 53.0 (#74)
- Qwen2.5 7B Instruct: —

| Benchmark | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Non-English | 1420 | — |
| LMArena Chinese | 1480 | — |
| LMArena French | 1461 | — |
| LMArena German | 1431 | — |
| LMArena Japanese | 1376 | — |
| LMArena Korean | 1397 | — |
| LMArena Russian | 1422 | — |
| LMArena Spanish | 1404 | — |

## Instruction Following

- Grok 4.6: 75.4 (#63)
- Qwen2.5 7B Instruct: 63.2 (#231)

| Benchmark | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| IFEval | — | 74.1% |
| LMArena Instruction Following | 1431 | — |

## Long Context

- Grok 4.6: 44.5 (#66)
- Qwen2.5 7B Instruct: —

| Benchmark | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Longer Query | 1454 | — |

## Writing & Preference

- Grok 4.6: 62.3 (#80)
- Qwen2.5 7B Instruct: 48.8 (#195)

| Benchmark | Grok 4.6 | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Text | 1428 | — |
| LMArena Creative Writing | 1428 | — |
| WildBench | — | 73.1% |
| LMArena Multi-Turn | 1425 | — |

## FAQ

### Is Grok 4.6 better than Qwen2.5 7B Instruct?

Grok 4.6 is the stronger model overall, scoring 56.9 to 29.0 on the Noometry Index. Qwen2.5 7B Instruct costs 9.8× 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 7B Instruct?

Qwen2.5 7B Instruct is cheaper. It lists at $0.17 per million input tokens and $0.70 per million output tokens; Grok 4.6 lists at $2 and $6.

### Is Grok 4.6 or Qwen2.5 7B Instruct better for coding?

Grok 4.6 scores higher on coding benchmarks: 58.5 versus 36.5 in the Noometry coding category.

### Which has the bigger context window?

Grok 4.6 does, with 500K tokens against 131K.

### How many benchmarks do Grok 4.6 and Qwen2.5 7B Instruct share?

6 benchmarks have published results for both models. Grok 4.6 has 49 scored results on Noometry and Qwen2.5 7B Instruct has 15.