Qwen3.5 397B A17B

Qwen · qwen/qwen3.5-397b-a17b

← Back to leaderboard

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...

Qwen3.5 397B A17B ranks #40 of 348 models in the default AI Benchmark Hub order (quality first, then cost). At $1.36 blended per million tokens, it is cheaper than 43% of models with published API pricing. Its 262,144-token context window is about the catalog median (262,144 tokens). A quality index of 65.3 places it at #40 of 41 models with an Arena Elo signal.

open weightsimagetexttext+image+video->textvideo

Context

Max context: 262144
Max output: 65536

Pricing

Input / 1M: 0.39
Output / 1M: 2.34
Blend / 1M: 1.36

Quality

Quality index: 65.3

Provider

Provider: Qwen
Moderated: no

Monthly cost at this blended rate

Estimated API spend if every token is billed at $1.36per million (the blend of this model's input and output prices). Real bills depend on the input/output mix.

Tokens / monthEst. cost (USD)
1,000,000$1.36
10,000,000$13.65
100,000,000$136.50

Related models

Same lab, similar price, or similar quality — useful next comparisons.

More from Qwen

Closest API price

Closest quality index