Alibaba (Qwen)

Qwen3.5 397B-A17B

Type
language model
Context
262K tokens
Max output
16K
Released
15 February 2026
API string
qwen3.5-397b-a17b
Weights
open

Prices by provider

Provider Input, $ per 1M Output, $ per 1M In our data since
EmpirioLabs AI $0.172 $1.032 31 Jul 2026
Merge Gateway $0.172 $1.032 31 Jul 2026
OrcaRouter $0.172 $1.032 31 Jul 2026
siliconflow-cn $0.29 $1.74 31 Jul 2026
vultr $0.3 $2 31 Jul 2026
crof $0.35 $1.75 31 Jul 2026
Kilo Code $0.39 $2.34 31 Jul 2026
OpenRouter qwen/qwen3.5-397b-a17b $0.39 $2.34 31 Jul 2026
SiliconFlow $0.39 $2.34 31 Jul 2026
alibaba-cn $0.43 $2.58 31 Jul 2026
DeepInfra $0.45 $3 31 Jul 2026
DigitalOcean Gradient AI $0.55 $3.5 31 Jul 2026
mixlayer $0.6 $3.6 31 Jul 2026
novita-ai $0.6 $3.6 31 Jul 2026
tensormesh $0.6 $3.6 31 Jul 2026
Alibaba (Qwen) $0.6 $3.6 31 Jul 2026
Cortecs $0.6 $3.6 31 Jul 2026
Hugging Face $0.6 $3.6 31 Jul 2026
NanoGPT $0.6 $3.6 31 Jul 2026
Nebius Token Factory $0.6 $3.6 31 Jul 2026
OpenRouter openrouter/qwen/qwen3.5-397b-a17b $0.6 $3.6 31 Jul 2026
Scaleway $0.6 $3.6 31 Jul 2026
Together AI $0.6 $3.6 31 Jul 2026
neuralwatt $0.69 $4.14 31 Jul 2026
OVHCloud AI Endpoints $0.71 $4.25 31 Jul 2026
Venice AI $0.75 $4.5 31 Jul 2026
greenpt $0.798 $4.959 31 Jul 2026

Only the base tier and only per-token prices are shown. Batch, discounted and cached rates, as well as prices per image or per second of video, do not go into this table: they cannot stand in the same column as a price per million tokens.

A single provider appears several times if it sells this model under different identifiers and at different prices — for example, at different weight precision. The identifier is shown next to the name. Identical offers that arrived from two sources under different spellings are merged into one row.

The date in the last column is the day this price first entered our collection. The price may well be older: before that day we simply were not recording it. It has not changed since — otherwise a new row with a new date would stand in its place.

Measurement results

Task set Result Run conditions Measured by
Arena Score, expert questions 1,467.1 1,459–1,476
Arena Score, programming 1,465.71 1,460–1,471
Arena Score in French 1,457.54 1,442–1,473
Arena Score in Spanish 1,449.69 1,434–1,465
Arena Score, hard prompts 1,449.17 1,445–1,454
Arena Score, mathematics 1,448.89 1,438–1,459
Arena Score, multi-turn dialogue 1,445.42 1,439–1,452
Arena Score in English 1,444.26 1,440–1,449
Arena Score, long queries 1,444.01 1,439–1,449
Arena Score, overall 1,439.09 1,435–1,443
Arena Score in Russian 1,432.76 1,425–1,441
Arena Score, instruction following 1,421.53 1,416–1,427
Arena Score, creative writing 1,404.33 1,397–1,411
Arena Score, web development 1,399.59 1,394–1,405
Arena Score, understanding diagrams 1,280.69 1,272–1,289
Arena Score, text recognition in images 1,275.27 1,269–1,281
Arena Score, working with images 1,266.03 1,260–1,272