# Grok 4 vs Qwen3.5-9B

> Grok 4 is the stronger model overall, scoring 47.8 to 33.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-4-vs-qwen3-5-9b
- Last updated: 2026-10-10
- Shared benchmarks: 5

## Summary

- They share 5 benchmarks with published results for both. Grok 4 scores higher in 5 categories and Qwen3.5-9B in 0 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Grok 4 leads 32.1 to 14.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 84% for Grok 4 and 61.7% for Qwen3.5-9B.
- Qwen3.5-9B has downloadable open weights; the other is API-only.

## Snapshot

| | Grok 4 | Qwen3.5-9B |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 47.8 | 33.3 |
| Rank | 49 | 235 |
| Context | — | 262K |
| Input $/M | — | $0.10 |
| Output $/M | — | $0.15 |
| Weights | Proprietary | Open |

## Coding

- Grok 4: 49.9 (#44)
- Qwen3.5-9B: 35.3 (#213)

| Benchmark | Grok 4 | Qwen3.5-9B |
|---|---|---|
| Aider Polyglot | 79.6% | — |
| SciCode | — | 27.5% |
| WeirdML | 45.7% | — |
| LMArena Coding | 1408 | — |

## Agentic & Tool Use

- Grok 4: 32.1 (#62)
- Qwen3.5-9B: 14.4 (#143)

| Benchmark | Grok 4 | Qwen3.5-9B |
|---|---|---|
| Terminal-Bench | 27.2% | 9.2% |
| Berkeley Function Calling Leaderboard | 63% | — |
| GDPval | 21.1% | — |
| Cybench | 43% | — |
| DeepResearch Bench | 47.3% | — |
| BALROG | 43.6% | — |
| LMArena Search | 1142 | — |
| METR Time Horizons | 66.6% | — |

## Reasoning

- Grok 4: 36.9 (#59)
- Qwen3.5-9B: 22.5 (#178)

| Benchmark | Grok 4 | Qwen3.5-9B |
|---|---|---|
| Chess Puzzles | 28% | 12% |
| Epoch Capabilities Index | 146.46 | 139.45 |
| ARC-AGI-2 | 16% | — |
| SimpleBench | 60.5% | — |
| Kagi LLM Benchmark | 73.6% | — |
| ARC-AGI-1 | 66.7% | — |
| CritPt | — | 0.3% |
| LMArena Hard Prompts | 1409 | — |
| DTBench | — | 71.2% |
| LMCA | — | 24.5% |
| ForecastBench | 60.9 | — |

## Math

- Grok 4: 46.5 (#58)
- Qwen3.5-9B: 33.5 (#174)

| Benchmark | Grok 4 | Qwen3.5-9B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 84% | 61.7% |
| MathArena Final-Answer Competitions | — | 48.5% |
| Omni-MATH | 60.3% | — |
| LMArena Math | 1422 | — |
| FrontierMath (Feb 2025 set) | 19.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- Grok 4: 53.7 (#50)
- Qwen3.5-9B: 46.0 (#77)

| Benchmark | Grok 4 | Qwen3.5-9B |
|---|---|---|
| GPQA Diamond | 87% | 79% |
| MMLU-Pro | 85.1% | — |
| Confabulations | 12.4% | — |
| GPQA (HELM) | 72.7% | — |
| LMArena Expert | 1417 | — |

## Multimodal

- Grok 4: 33.1 (#89)
- Qwen3.5-9B: —

| Benchmark | Grok 4 | Qwen3.5-9B |
|---|---|---|
| LMArena Vision | 1211 | — |
| GeoBench | 45% | — |

## Multilingual

- Grok 4: 52.0 (#98)
- Qwen3.5-9B: —

| Benchmark | Grok 4 | Qwen3.5-9B |
|---|---|---|
| LMArena Non-English | 1403 | — |
| LMArena Chinese | 1428 | — |
| LMArena French | 1418 | — |
| LMArena German | 1429 | — |
| LMArena Japanese | 1396 | — |
| LMArena Korean | 1376 | — |
| LMArena Russian | 1409 | — |
| LMArena Spanish | 1419 | — |

## Instruction Following

- Grok 4: 79.4 (#4)
- Qwen3.5-9B: —

| Benchmark | Grok 4 | Qwen3.5-9B |
|---|---|---|
| IFEval | 94.9% | — |
| LMArena Instruction Following | 1387 | — |

## Long Context

- Grok 4: 63.1 (#4)
- Qwen3.5-9B: —

| Benchmark | Grok 4 | Qwen3.5-9B |
|---|---|---|
| Fiction.LiveBench | 94.4% | — |
| LMArena Longer Query | 1409 | — |

## Writing & Preference

- Grok 4: 58.7 (#111)
- Qwen3.5-9B: —

| Benchmark | Grok 4 | Qwen3.5-9B |
|---|---|---|
| LMArena Text | 1412 | — |
| LMArena Creative Writing | 1397 | — |
| Short-Story Creative Writing | 76.9% | — |
| WildBench | 79.7% | — |
| LMArena Multi-Turn | 1416 | — |

## FAQ

### Is Grok 4 better than Qwen3.5-9B?

Grok 4 is the stronger model overall, scoring 47.8 to 33.3 on the Noometry Index.

### Is Grok 4 or Qwen3.5-9B better for coding?

Grok 4 scores higher on coding benchmarks: 49.9 versus 35.3 in the Noometry coding category.

### How many benchmarks do Grok 4 and Qwen3.5-9B share?

5 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Qwen3.5-9B has 10.