# Qwen3.5 35B-A3B vs Sonar

> Qwen3.5 35B-A3B is the stronger model overall, scoring 41.7 to 37.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen3-5-35b-a3b-vs-sonar
- Last updated: 2026-10-10
- Shared benchmarks: 0

## Summary

- The widest gap is in math, where Qwen3.5 35B-A3B leads 38.4 to 31.3.
- Qwen3.5 35B-A3B is cheaper at $0.25 / $2 per million input/output tokens, against $1 / $1 for Sonar.
- Qwen3.5 35B-A3B accepts more context: 262K tokens versus 128K.
- Qwen3.5 35B-A3B has downloadable open weights; the other is API-only.

## Snapshot

| | Qwen3.5 35B-A3B | Sonar |
|---|---|---|
| Provider | Alibaba (Qwen) | Perplexity |
| Noometry Index | 41.7 | 37.8 |
| Rank | 118 | 184 |
| Context | 262K | 128K |
| Input $/M | $0.25 | $1 |
| Output $/M | $2 | $1 |
| Weights | Open | Proprietary |

## Coding

- Qwen3.5 35B-A3B: 33.5 (#244)
- Sonar: 35.1 (#217)

| Benchmark | Qwen3.5 35B-A3B | Sonar |
|---|---|---|
| LMArena WebDev | 1254 | — |
| SciCode | 29.3% | — |
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1410 | — |

## Reasoning

- Qwen3.5 35B-A3B: 23.9 (#158)
- Sonar: 20.3 (#221)

| Benchmark | Qwen3.5 35B-A3B | Sonar |
|---|---|---|
| CritPt | 0.6% | — |
| Chess Puzzles | 10% | — |
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1400 | — |
| DTBench | 80% | — |
| LiveBench Data Analysis | — | 37.9% |
| LMCA | 29.5% | — |
| Epoch Capabilities Index | 142.52 | — |
| LiveBench | — | 46.9% |

## Math

- Qwen3.5 35B-A3B: 38.4 (#86)
- Sonar: 31.3 (#197)

| Benchmark | Qwen3.5 35B-A3B | Sonar |
|---|---|---|
| MathArena Final-Answer Competitions | 56% | — |
| OTIS Mock AIME 2024-2025 | 70% | — |
| LiveBench Math | — | 41.6% |
| LMArena Math | 1404 | — |

## Knowledge

- Qwen3.5 35B-A3B: 47.7 (#73)
- Sonar: —

| Benchmark | Qwen3.5 35B-A3B | Sonar |
|---|---|---|
| GPQA Diamond | 83.5% | — |
| Vectara Hallucination Rate | 10.5% | — |
| LMArena Expert | 1408 | — |

## Multilingual

- Qwen3.5 35B-A3B: 50.2 (#122)
- Sonar: —

| Benchmark | Qwen3.5 35B-A3B | Sonar |
|---|---|---|
| LMArena Non-English | 1379 | — |
| LMArena Chinese | 1458 | — |
| LMArena French | 1412 | — |
| LMArena German | 1367 | — |
| LMArena Japanese | 1326 | — |
| LMArena Korean | 1354 | — |
| LMArena Russian | 1377 | — |
| LMArena Spanish | 1390 | — |

## Instruction Following

- Qwen3.5 35B-A3B: 73.0 (#123)
- Sonar: 71.5 (#146)

| Benchmark | Qwen3.5 35B-A3B | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| LMArena Instruction Following | 1379 | — |

## Long Context

- Qwen3.5 35B-A3B: 42.4 (#122)
- Sonar: —

| Benchmark | Qwen3.5 35B-A3B | Sonar |
|---|---|---|
| LMArena Longer Query | 1390 | — |

## Writing & Preference

- Qwen3.5 35B-A3B: 58.1 (#118)
- Sonar: 52.6 (#166)

| Benchmark | Qwen3.5 35B-A3B | Sonar |
|---|---|---|
| LMArena Text | 1395 | — |
| LMArena Creative Writing | 1347 | — |
| LMArena Multi-Turn | 1391 | — |
| LiveBench Language | — | 44.1% |

## FAQ

### Is Qwen3.5 35B-A3B better than Sonar?

Qwen3.5 35B-A3B is the stronger model overall, scoring 41.7 to 37.8 on the Noometry Index.

### Which is cheaper, Qwen3.5 35B-A3B or Sonar?

Qwen3.5 35B-A3B is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Sonar lists at $1 and $1.

### Is Qwen3.5 35B-A3B or Sonar better for coding?

Sonar scores higher on coding benchmarks: 35.1 versus 33.5 in the Noometry coding category.

### Which has the bigger context window?

Qwen3.5 35B-A3B does, with 262K tokens against 128K.

### How many benchmarks do Qwen3.5 35B-A3B and Sonar share?

0 benchmarks have published results for both models. Qwen3.5 35B-A3B has 28 scored results on Noometry and Sonar has 7.