$0.2007
Estimated priceConnecting to model catalogs and market prices…
Qwen
Qwen3.5 397B A17B
Verified suppliersTrusted suppliers sell this model through GPU Router, and our marketplace orders only go to sellers vetted as trusted.qwen/qwen3.5-397b-a17bEst. 48% off · Standardvs cheapest OpenRouter endpoint for a typical request: ~80k input / 1k outputThe 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...
Standard prices and discounts apply to Standard requests. Requests requiring zero data retention use separate ZDR quotes. Retention depends on the provider serving your request.
$1.45
Estimated priceMarketplace price history.
Marketplace observations and reference prices.
Discount vs cheapest OpenRouter endpoint · typical request · higher is better
Loading recorded prices…
Cheapest OpenRouter endpoint now 0% off ($2.73)Our price (live offer) now 47.82% off ($1.65)
Higher is better. Each point prices a typical request of the model (~80k input / 1k output) at the recorded market rates, against today’s cheapest OpenRouter endpoint for the same request, using input and output rates only on both sides; OpenRouter history is not recorded. Negative values mean above the OpenRouter reference. Dashed lines show current references and the gateway estimate. Recorded market prices exclude the routing fee; gaps stay visible. In + out = 1M input + 1M output tokens.
Use this model
Qwen3.5 397B A17B qwen/qwen3.5-397b-a17b
- 01Set up GPU Router
Your coding assistant adds the client and checks model availability.
- 02Choose your calls
Switch all matching calls, selected features, or use it for new calls only.
- 03Switch and test
Only the calls you choose are migrated. Everything else stays as is.
Paste into your coding assistant. It will ask before switching calls.
What would your workload cost?
Change your token mix to compare the same workload.
Estimated savings with GPU Router$16.04less than the cheapest OpenRouter endpoint
For 80M input + 1M output tokens using our current estimate.
Our price is the cheapest eligible route or live trusted seller offer for this workload, including the routing fee. Final charges depend on the route and actual usage.