Qwen3.5-35B-A3B
Qwen released 2026-02-25
Gemma 4 31B is both better and cheaper.
It scores +5.4 higher and costs 68% less ($0.160/M against $0.500/M) on this workload — and it does everything this model does.
DeepSeek V4 Flash 0731 is cheaper still (79% less) but drops no image, video input.
Every price dimension
| Input | $0.250/M |
|---|---|
| Output | $1.25/M |
| Cached input | $0.250/M |
Independent scores
| Intelligence | 24.3 |
|---|---|
| Coding | 37 |
| Agentic | 11.8 |
| LMArena Elo | 1395.6default |
Sources are listed separately rather than averaged. Across the 62 models both have scored they correlate at r = 0.806 — close agreement overall, but four models rank very differently between them, and a blended score would hide exactly those.
Capability
| Context window | 262K |
|---|---|
| Max output | 262K |
| Input modes | text, image, video |
| Tool use | yes |
| Reasoning | optional |
Provenance
| Price source | openrouter.ai |
|---|---|
| Fetched | 2026-08-24 |
| Quality data | verified |
The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency. Its overall...