All Models
Qwen3.5 397B A17B
The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. It delivers...
Benchmarks
Available Providers (19)
| Provider | Model ID | Input Cost | Output Cost | Context | Max Output | Docs |
|---|---|---|---|---|---|---|
 Nvidia | qwen/qwen3.5-397b-a17b | $0/MTok | $0/MTok | 262.1K | 8.2K | |
 OrcaRouter | qwen/qwen3.5-397b-a17b | $0.17/MTok | $1.03/MTok | 262.1K | 65.5K | |
 CrofAI | qwen3.5-397b-a17b | $0.35/MTok | $1.75/MTok | 262.1K | 262.1K | |
 OpenRouter | qwen/qwen3.5-397b-a17b | $0.39/MTok | $2.45/MTok | 131.1K | 65.5K | |
 Kilo Gateway | qwen/qwen3.5-397b-a17b | $0.39/MTok | $2.34/MTok | 262.1K | 65.5K | |
 Alibaba (China) | qwen3.5-397b-a17b | $0.43/MTok | $2.58/MTok | 262.1K | 65.5K | |
 Wafer | Qwen3.5-397B-A17B | $0.43/MTok | $2.60/MTok | 262.1K | 65.5K | |
 Hugging Face | Qwen/Qwen3.5-397B-A17B | $0.60/MTok | $3.60/MTok | 262.1K | 32.8K | |
 NovitaAI | qwen/qwen3.5-397b-a17b | $0.60/MTok | $3.60/MTok | 262.1K | 64K | |
 Cortecs | qwen3.5-397b-a17b | $0.60/MTok | $3.60/MTok | 250K | 250K | |
 Nebius Token Factory | Qwen/Qwen3.5-397B-A17B | $0.60/MTok | $3.60/MTok | 262.1K | 8.2K | |
 Mixlayer | qwen/qwen3.5-397b-a17b | $0.60/MTok | $3.60/MTok | 262.1K | 262.1K | |
 LLM Gateway | qwen35-397b-a17b | $0.60/MTok | $3.60/MTok | 262.1K | 65.5K | |
 Scaleway | qwen3.5-397b-a17b | $0.60/MTok | $3.60/MTok | 256K | 16.4K | |
 Together AI | Qwen/Qwen3.5-397B-A17B | $0.60/MTok | $3.60/MTok | 262.1K | 130K | |
 Alibaba | qwen3.5-397b-a17b | $0.60/MTok | $3.60/MTok | 262.1K | 65.5K | |
 NanoGPT | qwen/qwen3.5-397b-a17b | $0.60/MTok | $3.60/MTok | 258.0K | 65.5K | |
 OVHcloud AI Endpoints | qwen3.5-397b-a17b | $0.71/MTok | $4.25/MTok | 262.1K | 262.1K | |
 Qiniu | qwen3.5-397b-a17b | —/MTok | —/MTok | 256K | 64K |
Capabilities
Reasoning
Tool Calling
Attachments
Open Weights
Structured Output