Qwen: Qwen3.5 397B A17B
qwen/qwen3.5-397b-a17b
73rd smartest of 182ReasoningToolsJSONVision
Intelligence
33.7
73rd of 182
Design Elo
1218
3D
Speed
—
tokens/sec
Latency
—
first token
Input price
$0.390
153rd cheapest
Context
262K
66K max out
How it compares
Smarter than60%
of all ranked models
Cheaper than54%
of all ranked models
Overview
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...
Benchmarks
independent · Artificial Analysis & Design ArenaArtificial Analysis81th percentile
Intelligence Index
33.7
Coding Index
48.2
GPQA Diamond
89%
Humanity's Last Exam
27%
SciCode
42%
Tau²-Bench (agentic)
96%
Design Arena · Elo960 tournaments
3D
1218
Website
1215
Data Viz
1204
svg
1189
Providers & pricing (11)
| Provider | In $/M | Out $/M | Context | Uptime |
|---|---|---|---|---|
| DigitalOcean | $0.385 | $2.45 | 131K | 98.1% |
| Alibabafp8 | $0.390 | $2.34 | 262K | 99.8% |
| Chutesfp8 | $0.450 | $3.00 | 262K | 100% |
| DeepInfrafp8 | $0.450 | $3.00 | 262K | 97.6% |
| Parasailfp8 | $0.500 | $3.60 | 262K | 99.9% |
| AtlasCloudfp8 | $0.550 | $3.50 | 262K | 99.5% |
| Phala | $0.550 | $3.50 | 262K | 99.2% |
| Novita | $0.600 | $3.60 | 262K | — |
| StreamLake | $0.600 | $3.60 | 256K | — |
| GMICloudfp8 | $0.600 | $3.60 | 262K | — |
| Venice | $0.750 | $4.50 | 128K | — |
Specifications
Context window262K
Max output66K
Knowledge cutoff—
Input modalitiestext, image, video
Output modalitiestext
Prompt caching—
Cache read price—
ModeratedNo
Open weightsQwen/Qwen3.5-397B-A17B ↗
Qwen3.5 397B A17B FAQ
How much does Qwen3.5 397B A17B cost?
Qwen3.5 397B A17B costs $0.390 per million input tokens and $2.34 per million output tokens via OpenRouter, making it 153rd cheapest of 332 paid models.
How smart is Qwen3.5 397B A17B?
Qwen3.5 397B A17B scores 33.7 on the Artificial Analysis Intelligence Index, ranking 73rd of 182 benchmarked models, with a GPQA Diamond score of 89%.
What is Qwen3.5 397B A17B's context window?
Qwen3.5 397B A17B supports a 262K-token context window and can output up to 66K tokens. It accepts text, image, video input.