Qwen: Qwen3.5-35B-A3B
by QwenQwen: Qwen3.5-35B-A3B is a large language model from Qwen. It costs $0.313 per million input tokens and $1.25 per million output tokens. Its context window is 262K tokens.
- Input / 1M tokens
- $0.313
- Output / 1M tokens
- $1.25
- Cached input / 1M
- $0.156
- Context window
- 262K
On repeated prefixes
tokens
Who serves it cheapest
7 hosts serve Qwen3.5-35B-A3B. Same weights, same API - the price difference is pure margin and routing.
| Provider | Input / 1M | Output / 1M | Context | Throughput | Uptime 24h |
|---|---|---|---|---|---|
| DarkbloomCheapestfp4 | $0.080 | $0.750 | 262K | - | 99.6% |
| DeepInfrafp8 | $0.140 | $1.00 | 262K | - | 99.8% |
| Parasailfp8 | $0.150 | $1.00 | 262K | - | 98.7% |
| Alibaba | $0.163 | $1.30 | 262K | - | 99.8% |
| Venice | $0.313 | $1.25 | 256K | - | 99.6% |
| AtlasCloudfp8 | $0.225 | $1.80 | 262K | - | 99.3% |
| SiliconFlowfp8 | $0.240 | $1.80 | 262K | - | 99.2% |
The spread between Darkbloom and SiliconFlow is 2.5× for identical weights. Quantization and context limits differ, so check both columns before switching.
Benchmarks
Independent scores published alongside the catalogue.
- Coding index
- 37.0
About Qwen3.5-35B-A3B
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...
Specifications
| Model ID | qwen/qwen3.5-35b-a3b |
|---|---|
| Provider | Qwen |
| Context window | 262K tokens |
| Max output | 16K tokens |
| Input modalities | text, image, video |
| Output modalities | text |
| Knowledge cutoff | - |
| Open weights | Yes - Qwen/Qwen3.5-35B-A3B |
| Released | February 25, 2026 |
Cheaper alternatives
Models that cost less than Qwen3.5-35B-A3B while keeping at least half its context window and every input modality it supports.
Frequently asked
How much does Qwen: Qwen3.5-35B-A3B cost?
$0.313 per million input tokens and $1.25 per million output tokens. Cached input reads cost $0.156 per million tokens.
What is the context window of Qwen: Qwen3.5-35B-A3B?
262K tokens, with up to 16K tokens of output per request.
Which provider serves Qwen: Qwen3.5-35B-A3B cheapest?
Darkbloom at $0.080 per million input tokens - 2.5× cheaper than SiliconFlow, the most expensive of the 7 hosts serving it.
Confirm against the source: Qwen official pricing.