Qwen3.8 Flash

Name: Qwen3.8 Flash
Author: Qwen (Alibaba)

by Qwen (Alibaba)

APIAvailable through a hosted API — pay per token, no self-hosting required

Released August 20261000K context≈ 750,000 words125000000000 params

Qwen3.8 Flash is Alibaba's cost-efficient entry in the Qwen 3.8 family, using a mixture-of-experts architecture with 125 billion total parameters but only 6 billion active per forward pass. This MoE design delivers strong performance at a fraction of the compute cost of dense models. It handles both text and image inputs with a 1M-token context window. At $0.15 input and $0.47 output per million tokens, it sits among the cheapest multimodal models available via API, positioned below the heavier Qwen3.8 Max for latency-sensitive and high-volume workloads.

Qwen3.8 Flash

by Qwen (Alibaba)

APIAvailable through a hosted API — pay per token, no self-hosting required

Released August 20261000K context≈ 750,000 words125000000000 params

Qwen3.8 Flash is Alibaba's cost-efficient entry in the Qwen 3.8 family, using a mixture-of-experts architecture with 125 billion total parameters but only 6 billion active per forward pass. This MoE design delivers strong performance at a fraction of the compute cost of dense models. It handles both text and image inputs with a 1M-token context window. At $0.15 input and $0.47 output per million tokens, it sits among the cheapest multimodal models available via API, positioned below the heavier Qwen3.8 Max for latency-sensitive and high-volume workloads.

Glossary

APIAn interface that allows developers to send requests to and receive responses from an AI model over the internet.ArchitectureThe underlying structural design of a neural network that defines how data flows through layers and components.Context WindowThe maximum number of tokens a model can process in a single conversation or prompt.Forward PassA single computation cycle where input data flows through the model's layers to produce an output prediction.LatencyThe time delay between sending a request and receiving the first response token from a model.MultimodalA model that can process and understand multiple types of input, such as both text and images.ParametersThe learned numerical values in a model — more parameters generally means more capacity but higher compute cost.TokenA small unit of text (a word, subword, or punctuation mark) that a language model breaks input into for processing.TokensThe basic units of text that a language model processes, typically representing words or word fragments.