Qwen3.5-35B-A3B pricing & benchmarks
Qwen3.5-35B-A3B is available from Alibaba at $0.140 per million input tokens and $1.00 per million output tokens ($0.355 blended at 3:1). It accepts up to 262,144 tokens of context. 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...
Pricing
Per million tokens · OpenRouter
Input$0.140
Output$1.00
Blended (3:1)$0.355
Specification
qwen/qwen3.5-35b-a3b
Context window262K
Max output82K
ReleasedFeb 2026
LicenceOpen weights
Modalitiesin:text, in:image, in:video, out:text
No independently-run benchmark scores are published for Qwen3.5-35B-A3B yet. We show nothing rather than reprinting an unverified figure.
Where to run Qwen3.5-35B-A3B
7 hosts · 2.7× between cheapest and dearest. Blended at 3:1. ⚠ marks an option serving less context than the best available, or re-pricing above a prompt-length threshold. Bracketed labels are the provider's own service tier or region.
| Host | Blended $/1M | In / Out | Context | Weights | Uptime 24h |
|---|---|---|---|---|---|
| Darkbloom | $0.236 | $0.065 / $0.750 | 262K | fp4 | 99.6% |
| DeepInfra | $0.355 | $0.140 / $1.00 | 262K | fp8 | 99.8% |
| Parasail | $0.362 | $0.150 / $1.00 | 262K | fp8 | 100.0% |
| Venice | $0.362 | $0.150 / $1.00 | 256K | — | 100.0% |
| Alibaba | $0.447 | $0.163 / $1.30 | 262K | — | 99.8% |
| AtlasCloud | $0.619 | $0.225 / $1.80 | 262K | fp8 | 100.0% |
| SiliconFlow | $0.630 | $0.240 / $1.80 | 262K | fp8 | 85.7% |
Host prices and uptime from OpenRouter, synced 2026-10-22. Lower-precision weights (fp4, fp8) can cost less and score worse than the same model at bf16 — the score above is not host-specific.
Compare Qwen3.5-35B-A3B
FAQ
How much does Qwen3.5-35B-A3B cost?
Qwen3.5-35B-A3B costs $0.140 per million input tokens and $1.00 per million output tokens on OpenRouter. At a 3:1 input:output mix that blends to $0.355 per million tokens.
What is Qwen3.5-35B-A3B's context window?
Qwen3.5-35B-A3B accepts up to 262,144 tokens of context, and can generate up to 81,920 output tokens.