# Grok 3 vs Qwen3.5 27B

> Qwen3.5 27B is the stronger model overall, scoring 41.9 to 39.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-3-vs-qwen3-5-27b
- Last updated: 2026-10-12
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Grok 3 scores higher in 4 categories and Qwen3.5 27B in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.5 27B leads 27.5 to 13.7.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 5.8% for Grok 3 and 12.1% for Qwen3.5 27B.
- Qwen3.5 27B has downloadable open weights; the other is API-only.

## Snapshot

| | Grok 3 | Qwen3.5 27B |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 39.9 | 41.9 |
| Rank | 157 | 127 |
| Context | — | 262K |
| Input $/M | — | $0.30 |
| Output $/M | — | $2.40 |
| Weights | Proprietary | Open |

## Coding

- Grok 3: 41.9 (#115)
- Qwen3.5 27B: 38.9 (#168)

| Benchmark | Grok 3 | Qwen3.5 27B |
|---|---|---|
| WeirdML | 37.2% | 39.5% |
| LMArena Coding | 1432 | 1427 |
| Aider Polyglot | 53.3% | — |
| LMArena WebDev | — | 1358 |
| ALE-Bench | — | 349.45 |

## Agentic & Tool Use

- Grok 3: 30.5 (#76)
- Qwen3.5 27B: —

| Benchmark | Grok 3 | Qwen3.5 27B |
|---|---|---|
| BALROG | 29.5% | — |
| Vending-Bench 2 | — | 201.98 |

## Reasoning

- Grok 3: 13.7 (#333)
- Qwen3.5 27B: 27.5 (#117)

| Benchmark | Grok 3 | Qwen3.5 27B |
|---|---|---|
| LMArena Hard Prompts | 1434 | 1414 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 36.1% | — |
| Kagi LLM Benchmark | 61.3% | — |
| NYT Connections (extended) | — | 47.9% |
| ARC-AGI-1 | 5.5% | — |
| Thematic Generalization | — | 45.5% |
| DTBench | — | 82.4% |
| LMCA | — | 34% |
| Epoch Capabilities Index | 138.33 | — |

## Math

- Grok 3: 38.0 (#145)
- Qwen3.5 27B: 38.8 (#127)

| Benchmark | Grok 3 | Qwen3.5 27B |
|---|---|---|
| LMArena Math | 1391 | 1429 |
| MathArena Final-Answer Competitions | — | 56.7% |
| OTIS Mock AIME 2024-2025 | 55.6% | — |
| Omni-MATH | 46.4% | — |
| MATH Level 5 | 88.7% | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |

## Knowledge

- Grok 3: 46.2 (#82)
- Qwen3.5 27B: 38.0 (#150)

| Benchmark | Grok 3 | Qwen3.5 27B |
|---|---|---|
| Vectara Hallucination Rate | 5.8% | 12.1% |
| LMArena Expert | 1421 | 1428 |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 78.8% | — |
| Confabulations | 14.2% | — |
| GPQA (HELM) | 65% | — |

## Multimodal

- Grok 3: —
- Qwen3.5 27B: 39.4 (#59)

| Benchmark | Grok 3 | Qwen3.5 27B |
|---|---|---|
| LMArena Vision | — | 1241 |

## Multilingual

- Grok 3: 52.3 (#87)
- Qwen3.5 27B: 50.8 (#115)

| Benchmark | Grok 3 | Qwen3.5 27B |
|---|---|---|
| LMArena Non-English | 1410 | 1390 |
| LMArena Chinese | 1448 | 1478 |
| LMArena French | 1460 | 1410 |
| LMArena German | 1431 | 1393 |
| LMArena Japanese | 1387 | 1345 |
| LMArena Korean | 1373 | 1358 |
| LMArena Russian | 1416 | 1390 |
| LMArena Spanish | 1417 | 1407 |

## Instruction Following

- Grok 3: 75.0 (#73)
- Qwen3.5 27B: 73.5 (#119)

| Benchmark | Grok 3 | Qwen3.5 27B |
|---|---|---|
| LMArena Instruction Following | 1409 | 1393 |
| IFEval | 88.4% | — |

## Long Context

- Grok 3: 38.7 (#192)
- Qwen3.5 27B: 43.1 (#106)

| Benchmark | Grok 3 | Qwen3.5 27B |
|---|---|---|
| LMArena Longer Query | 1439 | 1413 |
| Fiction.LiveBench | 58.3% | — |

## Writing & Preference

- Grok 3: 55.8 (#141)
- Qwen3.5 27B: 59.3 (#111)

| Benchmark | Grok 3 | Qwen3.5 27B |
|---|---|---|
| LMArena Text | 1426 | 1409 |
| LMArena Creative Writing | 1414 | 1362 |
| LMArena Multi-Turn | 1425 | 1410 |
| Short-Story Creative Writing | 76.4% | — |
| EQ-Bench Creative Writing | 1186 | — |
| WildBench | 84.9% | — |

## FAQ

### Is Grok 3 better than Qwen3.5 27B?

Qwen3.5 27B is the stronger model overall, scoring 41.9 to 39.9 on the Noometry Index.

### Is Grok 3 or Qwen3.5 27B better for coding?

Grok 3 scores higher on coding benchmarks: 41.9 versus 38.9 in the Noometry coding category.

### How many benchmarks do Grok 3 and Qwen3.5 27B share?

19 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Qwen3.5 27B has 28.