Qwen models
Which Qwen models can you run on Modal?
Qwen Coming soon
Qwen4-Max
Qwen's first Qwen4 model: a natively multimodal 2.4T-parameter flagship with 71B active per token and a 1M-token context, built for coding and long-horizon agentic work.
Text VisionQwen
Qwen3.8-Max
The open-weights, text-only Max-class model for coding and long-horizon agentic work.
TextQwen
Qwen3.8 27B
Qwen's compact dense Qwen3.8 model: a 27B model with switchable thinking, 256K context, and Apache 2.0 weights.
TextQwen
Qwen3.8 Flash Next FP8
Qwen's open preview of the Qwen4 architecture: a 125B MoE with 6B active, Qwen Sparse Attention, 51B n-gram embeddings, and 256K context, in FP8.
Text VisionQwen
Qwen3.8 2.4T A95B NVFP4
Qwen's open Qwen-Max-class flagship: a text-only 2.4T-parameter MoE with 95B active per token, always-on reasoning, and 256K context, quantized to NVFP4 by Modal.
TextQwen
Qwen3.6 35B A3B
Qwen's first open-weight Qwen3.6 model: a 35B MoE vision-language model with 3B active parameters, default-on thinking, 262K context, and Apache 2.0 weights.
Text Vision
| Model | Architecture | Context | Modalities | Input / Output |
|---|---|---|---|---|
| Qwen4-Max | 2.4T MoE · 71B active | 1M | Text · Image | Coming soon |
| Qwen3.8-Max | 2.4T MoE | Up to 1M | Text | $2.00 / $6.00 |
| Qwen3.8 27B | 27B dense | 256K | Text | Dedicated only |
| Qwen3.8 Flash Next FP8 | 125B MoE · 6B active | 256K | Text · Image | Dedicated only |
| Qwen3.8 2.4T A95B NVFP4 | 2.4T MoE · 95B active | 256K | Text | Dedicated only |
| Qwen3.6 35B A3B | 35B MoE · 3B active | 262K | Text · Image | Dedicated only |
| Qwen3.6 35B A3B FP8 | 35B MoE · 3B active | 262K | Text · Image | Dedicated only |
| Qwen3.6 27B | 27B dense | 262K | Text · Image | Dedicated only |
| Qwen3.6 27B FP8 | 27B dense | 262K | Text · Image | Dedicated only |
| Qwen3.5 9B | 9B dense | 256K | Text · Image | Dedicated only |
| Qwen3.5 4B | 4B dense | 256K | Text · Image | Dedicated only |
| Qwen3.5 2B | 2B dense | 256K | Text · Image | Dedicated only |
| Qwen3.5 0.8B | 0.8B dense | 256K | Text · Image | Dedicated only |
| Qwen3.5 397B A17B FP8 | 397B MoE · 17B active | 256K | Text · Image | Dedicated only |
| Qwen3.5 122B A10B FP8 | 122B MoE · 10B active | 256K | Text · Image | Dedicated only |
| Qwen3.5 35B A3B FP8 | 35B MoE · 3B active | 256K | Text · Image | Dedicated only |
| Qwen3.5 27B FP8 | 27B dense | 256K | Text · Image | Dedicated only |
| Qwen3 Embedding 8B | 8B dense | 32K | Text | Dedicated only |
| Qwen3 Embedding 0.6B | 0.6B dense | 32K | Text | Dedicated only |
| Qwen3.8-Max VL | 2.4T parameters | 1M | Text · Image | $2.00 / $6.00 |
All prices per million tokens. Dedicated Endpoint compute is billed by the second. See compute pricing.
Latest updates

Scaling Intelligence: A landmark commission and exhibition

Sandbox resourcing: burst, don't reserve
Why you're paying for capacity you aren't using, and how to pay only for what you use.

Modal becomes a gVisor maintainer
Modal runs upwards of 40 million gVisor containers every day. Here's why we bet on it, what we've built into it, and where we want to take it next.
