AI Observatory / Model Radar Qwen / qwen/qwen3.5-397b-a17b

Qwen: 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...

262,144 Context
65,536 Max output
$0.39 / 1M Prompt price
2026-02-16 Created
01 / Snapshot

Pricing, context, modalities, and parameters.

Model Radar detail pages stay neutral and operator-readable: core metadata first, then workflow fit.

Provider Qwen Input modalities text, image, video
Output modalities text Prompt price $0.39 / 1M
Completion price $2.34 / 1M Request price N/A
Context length 262,144 Max completion tokens 65,536
Supported parameters frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p
Best for qwen/qwen3.5-397b-a17b

Qwen: 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...

Cross-modal work Long context High-volume usage Coding workflows new vision tool-capable long-context low-cost
03 / Colophon

Routes and exits.

Each model page stays simple: overview, compare, related models, then back to the public hub.