Qwen3.5 397B-A17B
Summary
Qwen3.5 397B-A17B is an open-weight language model from Alibaba (Qwen), released on 15 Feb 2026. It accepts text, images and video and generates text, with a 262K-token context window and up to 66K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.45 per million input tokens and $3 per million output tokens (DeepInfra), across 5 providers we track.
Large open Qwen multimodal MoE for visual agents and long technical tasks
Last verified 18 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | Alibaba (Qwen) |
|---|---|
| Release date | 15 Feb 2026 [models.dev] |
| Status | Generally available [models.dev] |
| Weights | Open weights [Hugging Face] |
| License | apache-2.0 [Hugging Face] |
| Parameters | 403B [Hugging Face] |
| Context window | 262K (262,144 tokens) [models.dev] |
| Max output | 66K tokens [models.dev] |
| Knowledge cutoff | Apr 2025 [models.dev] |
| Input | Text, images and video [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | Yes [models.dev] |
| Tool calling | Yes [models.dev] |
| Structured output | Yes [models.dev] |
| Hugging Face | Qwen/Qwen3.5-397B-A17B [Hugging Face] |
Qwen3.5 397B-A17B API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| DeepInfra Qwen/Qwen3.5-397B-A17B | $0.45 | $3 | $0.22 | 262K | models.dev, 18 Nov 2026 |
| OpenRouter qwen/qwen3.5-397b-a17b | $0.45 | $3 | $0.22 | 262K | OpenRouter, 18 Nov 2026 |
| Alibaba Cloud Model Studio qwen3.5-397b-a17b | $0.60 | $3.60 | – | 262K | models.dev, 18 Nov 2026 |
| Hugging Face Inference Providers Qwen/Qwen3.5-397B-A17B | $0.60 | $3.60 | – | 262K | models.dev, 18 Nov 2026 |
| Novita AI qwen/qwen3.5-397b-a17b | $0.60 | $3.60 | – | 262K | models.dev, 18 Nov 2026 |
Running Qwen3.5 397B-A17B locally
Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 245 GB | 246 GB | 246 GB |
| Q8_0 (8-bit) | 429 GB | 430 GB | 430 GB |
| FP16 / BF16 (16-bit) | 807 GB | 808 GB | 809 GB |
At 4-bit with an 8K conversation, Qwen3.5 397B-A17B needs about 246 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.
Frequently asked questions
What is the context window of Qwen3.5 397B-A17B?
Qwen3.5 397B-A17B has a context window of 262,144 tokens (262K) and can generate up to 65,536 tokens in a single response, according to models.dev.
How much does the Qwen3.5 397B-A17B API cost?
Through Alibaba Cloud Model Studio, Qwen3.5 397B-A17B costs $0.60 per million input tokens and $3.60 per million output tokens. The lowest listed price is $0.45 input / $3 output through DeepInfra. Prices last verified 18 Nov 2026.
Is Qwen3.5 397B-A17B open source?
Qwen3.5 397B-A17B is an open-weight model: its weights are published on Hugging Face as Qwen/Qwen3.5-397B-A17B under the apache-2.0 license. Check the license terms before commercial use.
When was Qwen3.5 397B-A17B released?
Alibaba (Qwen) released Qwen3.5 397B-A17B on 15 Feb 2026, according to models.dev.
Which providers offer Qwen3.5 397B-A17B?
We track Qwen3.5 397B-A17B on 5 providers: DeepInfra, OpenRouter, Alibaba Cloud Model Studio, Hugging Face Inference Providers, Novita AI.
How much VRAM does Qwen3.5 397B-A17B need?
About 246 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 430 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count and architecture.
Similar models from other labs
Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| GLM-5 | Zhipu AI (Z.ai) | 12 Feb 2026 | 205K | $0.95 | $2.55 | Open |
| Kimi K2.5 | Moonshot AI | 27 Jan 2026 | 262K | $0.45 | $2.25 | Open |
| Kimi K2.6 | Moonshot AI | 20 Apr 2026 | 262K | $0.65 | $3.40 | Open |
| GLM-5.1 | Zhipu AI (Z.ai) | 7 Apr 2026 | 200K | $1 | $3.20 | Open |
| Inkling | Thinkingmachines | 15 Jul 2026 | 256K | $1 | $4.05 | Open |
| Mistral Medium 3.5 | Mistral AI | 29 Apr 2026 | 262K | $1.50 | $7.50 | Open |
Change history
- OpenRouter price changed: input $0.60 → $0.45, output $3.60 → $3 per 1M tokens
- Knowledge cutoff changed from Jan 2025 to Apr 2025
- No longer listed on nebius
- Knowledge cutoff changed from Apr 2025 to Jan 2025
- Knowledge cutoff changed from Jul 2025 to Apr 2025
- Knowledge cutoff changed from Jan 2025 to Jul 2025
- OpenRouter price changed: input $0.50 → $0.60, output $3.60 → $3.60 per 1M tokens
- Knowledge cutoff changed from Jul 2025 to Jan 2025
- Knowledge cutoff changed from Apr 2025 to Jul 2025
- Knowledge cutoff changed from Jul 2025 to Apr 2025
- Knowledge cutoff changed from Apr 2025 to Jul 2025
- Knowledge cutoff changed from Jan 2025 to Apr 2025
- Knowledge cutoff changed from Apr 2025 to Jan 2025
- Knowledge cutoff changed from Jan 2025 to Apr 2025
- Knowledge cutoff changed from Apr 2025 to Jan 2025
- Knowledge cutoff changed from Jan 2025 to Apr 2025
- Knowledge cutoff changed from Apr 2025 to Jan 2025
- nebius price changed: input $0.60 → $0.60, output $3.60 → $3.60 per 1M tokens
- OpenRouter price changed: input $0.39 → $0.50, output $2.34 → $3.60 per 1M tokens
- Knowledge cutoff changed from Jan 2025 to Apr 2025
- No longer listed on siliconflow
- nebius price changed: input $0.60 → $0.60, output $3.60 → $3.60 per 1M tokens
- Knowledge cutoff changed from Jan 2026 to Jan 2025
- OpenRouter price changed: input $0.50 → $0.39, output $3.60 → $2.34 per 1M tokens
- Input modalities changed from text, image, video, audio to text, image, video
Sources
- https://openrouter.ai/qwen/qwen3.5-397b-a17b
- https://github.com/sst/models.dev/blob/dev/providers/alibaba/models/qwen3.5-397b-a17b.toml
- https://huggingface.co/Qwen/Qwen3.5-397B-A17B
- https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/Qwen/Qwen3.5-397B-A17B.toml
- https://github.com/sst/models.dev/blob/dev/providers/huggingface/models/Qwen/Qwen3.5-397B-A17B.toml
- https://github.com/sst/models.dev/blob/dev/providers/novita-ai/models/qwen/qwen3.5-397b-a17b.toml