Skip to content
A new frontier for open-weight intelligence: Kimi K3.1 Pro is now available on Modal.
Learn more

Qwen models

Which Qwen models can you run on Modal?

ModelArchitectureContextModalitiesInput / Output
Qwen4-Max2.4T MoE · 71B active1MText · ImageComing soon
Qwen3.8-Max2.4T MoEUp to 1MText$2.00 / $6.00
Qwen3.8 27B27B dense256KTextDedicated only
Qwen3.8 Flash Next FP8125B MoE · 6B active256KText · ImageDedicated only
Qwen3.8 2.4T A95B NVFP42.4T MoE · 95B active256KTextDedicated only
Qwen3.6 35B A3B35B MoE · 3B active262KText · ImageDedicated only
Qwen3.6 35B A3B FP835B MoE · 3B active262KText · ImageDedicated only
Qwen3.6 27B27B dense262KText · ImageDedicated only
Qwen3.6 27B FP827B dense262KText · ImageDedicated only
Qwen3.5 9B9B dense256KText · ImageDedicated only
Qwen3.5 4B4B dense256KText · ImageDedicated only
Qwen3.5 2B2B dense256KText · ImageDedicated only
Qwen3.5 0.8B0.8B dense256KText · ImageDedicated only
Qwen3.5 397B A17B FP8397B MoE · 17B active256KText · ImageDedicated only
Qwen3.5 122B A10B FP8122B MoE · 10B active256KText · ImageDedicated only
Qwen3.5 35B A3B FP835B MoE · 3B active256KText · ImageDedicated only
Qwen3.5 27B FP827B dense256KText · ImageDedicated only
Qwen3 Embedding 8B8B dense32KTextDedicated only
Qwen3 Embedding 0.6B0.6B dense32KTextDedicated only
Qwen3.8-Max VL2.4T parameters1MText · Image$2.00 / $6.00

All prices per million tokens. Dedicated Endpoint compute is billed by the second. See compute pricing.

Ship your first app in minutes

$30 / month free compute