A
Qwen3.5 397B A17BOPEN
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.
Prices update automatically — checked daily against provider list prices.
- $0.88 per 1M tokens, assuming 3 input for 1 output the convention used by Artificial Analysis
- cheaper than 21% of the 191 open-weight models tracked here
- 41 of 51 Alibaba models tracked here, from cheapest
Benchmarks & performance
Independent benchmark scores for Qwen3.5 397B A17B, measured by Artificial Analysis. Higher is better. Measurement mode: Reasoning.
For the same budget, Qwen3.5 397B A17B delivers 4 times more capability. Open weights also mean you can self-host it and pay nothing per token.
Models at this level cost $1.56 per 1M tokens (median of 21) — this one costs $0.88.
Cheaper and better on this index: GLM 5.3 Flash · Gemini 3.7 Flash · Gemini 3.7 Flash (batch) and 28 more
About this model
Qwen3.5 397B A17B is an open-weight AI model by Alibaba. You can download and self-host it for free; the prices below are hosted-API list prices, tracked daily, for when you prefer convenience over self-hosting.
Frequently asked questions
What is Qwen3.5 397B A17B?
Qwen3.5 397B A17B is an AI language model from Alibaba. It is open-weight: you can download it and run it on your own hardware, for free. It scores 34.3 on the Artificial Analysis intelligence index.
Is Qwen3.5 397B A17B free?
The weights are free and open — you can self-host Qwen3.5 397B A17B and pay nothing per token. If you prefer a hosted API, list prices are $0.39 per million input tokens and $2.34 per million output tokens.
What is Qwen3.5 397B A17B good at?
Independent benchmarks from Artificial Analysis give it GPQA 89.3%, Humanity's Last Exam 29%, Long Context Reasoning 72.7%, SciCode 42%, IFBench 78.8%, τ²-Bench 95.6%, τ-Bench Banking 13.4%, Terminal-Bench 51.3%, Terminal-Bench Hard 40.9%. It is particularly used for code generation.
How fast is Qwen3.5 397B A17B?
It generates about 88.8 tokens per second, with a median 1.45s delay before the first token. Measured independently by Artificial Analysis.
Can I self-host Qwen3.5 397B A17B?
Yes. Qwen3.5 397B A17B has open weights, so you can download it and run it on your own GPU or server with tools like Ollama, vLLM or llama.cpp — with no per-token cost.