Qwen3.5 397B A17B
Qwen vision-language model for visual reasoning, documents, and agent tasks
Cheapest input
$0.172
$/MTok · models.dev
Cheapest output
$1.03
$/MTok
Context
262K
tokens
Offers
16
3 sources
Offers
Same model, different listings. First-party, router, and aggregator rows stay separate so you can see the spread instead of a single blended number. Free/$0 listings are hidden by default (1 omitted). Show free/$0 offers.
| Source | Provider | Input | Output | Cache read | Context |
|---|---|---|---|---|---|
| models.devcheapestqwen/qwen3.5-397b-a17b | Merge Gateway | $0.172 | $1.03 | $0.0344 | 256K |
| models.devqwen/qwen3.5-397b-a17b | OrcaRouter | $0.172 | $1.03 | $0 | 262K |
| models.devQwen/Qwen3.5-397B-A17B | SiliconFlow (China) | $0.29 | $1.74 | $0 | 262K |
| models.devQwen/Qwen3.5-397B-A17B | Vultr | $0.3 | $2 | $0 | 262K |
| models.devqwen/qwen3.5-397b-a17b | Kilo Gateway | $0.39 | $2.34 | $0 | 262K |
| OpenRouterqwen/qwen3.5-397b-a17b | OpenRouter | $0.39 | $2.34 | $0 | 262K |
| models.devQwen/Qwen3.5-397B-A17B | SiliconFlow | $0.39 | $2.34 | $0 | 262K |
| models.devQwen/Qwen3.5-397B-A17B | DeepInfra | $0.45 | $3 | $0.22 | 262K |
| models.devQwen/Qwen3.5-397B-A17B | Hugging Face | $0.6 | $3.60 | $0 | 262K |
| models.devqwen/qwen3.5-397b-a17b | Mixlayer | $0.6 | $3.60 | $0 | 262K |
| models.devqwen/qwen3.5-397b-a17b | NanoGPT | $0.6 | $3.60 | $0 | 258K |
| models.devQwen/Qwen3.5-397B-A17B | Nebius | $0.6 | $3.60 | $0.06 | 262K |
| models.devqwen/qwen3.5-397b-a17b | Novita AI | $0.6 | $3.60 | $0 | 262K |
| models.devQwen/Qwen3.5-397B-A17B | Together AI | $0.6 | $3.60 | $0.35 | 262K |
| LiteLLMQwen/Qwen3.5-397B-A17B | — | $0.6 | $3.60 | $0.06 | — |
| LiteLLMopenrouter/qwen/qwen3.5-397b-a17b | — | $0.6 | $3.60 | $0.06 | — |
Capabilities
- Reasoning
- Tool calling
- Structured output
- Attachments
- Open weights
- In: text
- In: image
- In: video
- In: audio
- Out: text
Knowledge cutoff: 2025-01Released: 2026-02-01Family: qwen
Scores
Sparse coverage on purpose. Missing a score means we have not linked one yet, not that the model is unranked. Full boards on /benchmarks.
- AA intelligence33.7
- AA coding48.2
- AA agentic19.8
Sources: Arena AI, Artificial Analysis. See /benchmarks.