Qwen
Qwen3.5 397B A17B
Launched 2026-02-15 · Listed on OpenRouter 2026-02-16
Prices updated . Benchmarks updated .
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...
You send / 1M
$0.450
tokens in
It replies / 1M
$3.00
tokens out
Context
258K
tokens it can read at once
Typical request
$0.0021
1,000 words in, 400 out
Benchmarks
| Benchmark | Score | Kind | Date | Source |
|---|---|---|---|---|
| Epoch Capabilities Index | 146.8 | Independent | 2026-02-13 | Epoch AI |
| GPQA Diamond | 86.4% | Independent | 2026-08-07 | Epoch AI |
| OTIS Mock AIME | 88.9% | Independent | 2026-08-07 | Epoch AI |
| CritPt | 0.9% | Independent | 2026-02-13 | CritPt, republished by Epoch AI |
| Arena Elo (text) | 1,438 | Independent | 2026-10-28 | LMArena |
| Arena Elo (webdev) | 1,400 | Independent | 2026-10-28 | LMArena |
| Arena Elo (vision) | 1,262 | Independent | 2026-10-19 | LMArena |
Rankings on this site use the Epoch Capabilities Index, an independent composite. Lab-reported rows are the lab's own figures and are kept separate.
Data privacy
No training or retention: Yes. At least one provider serves it with zero data retention.
| Provider | You send / 1M | It replies / 1M | Uptime | Privacy policy |
|---|---|---|---|---|
| DeepInfra · fp8 | $0.360 | $2.40 | 100% | Read |
| DeepInfra · fp8 | $0.450 | $3.00 | 94% | Read |
| Parasail · fp8 | $0.500 | $3.60 | 96% | Read |
| Novita | $0.600 | $3.60 | 99% | Read |
| DeepInfra · fp8 | $0.675 | $4.50 | 69% | Read |
| Venice | $0.750 | $4.50 | 99% | Read |
Based on OpenRouter and Vercel AI Gateway data. Confirm terms with the provider before sending sensitive data.
What it can do
- Reads: text, image, video
- Writes: text
- Weights: Open weights
- Reasoning: Yes
- Tool calling: Yes
- Longest reply: 64K tokens
- Repeated text is billed at $0.220 per 1M (prompt caching)
Compare with
Appears in
Estimate your monthly cost with Qwen3.5 397B A17B
Sources and licenses
- OpenRouter: Prices, context windows, modalities and the model list come from the OpenRouter public API. Data: OpenRouter.
- Vercel AI Gateway: The no-training flag comes from the Vercel AI Gateway model list.
- models.dev: Lab launch dates, open-weights flags and knowledge cutoffs come from models.dev. Copyright (c) 2025 models.dev, MIT License.
- Epoch AI: Independent benchmark scores and the Epoch Capabilities Index (ECI) come from Epoch AI's Benchmarking Hub, used under CC BY 4.0. Epoch AI, Capabilities & benchmarking, epoch.ai/benchmarks. Scores Epoch marks as external keep their original license, and the source is named on each score.
- LMArena: Arena Elo (text, webdev and vision) comes from the LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), CC BY 4.0, leaderboard date 2026-10-28.
- Lab-reported scores, when an editor has added them, are the lab's own numbers. Each one links to the source and is labeled lab-reported. They are not used to rank the lists.
- What a score measures: the benchmark guide.
