---
id: "qwen/qwen3.8-flash"
also: ["qwen/qwen3.8-flash-20260826", "Qwen/Qwen3.8-Flash-Next", "requesty/alibaba/qwen3.8-flash", "vercel/alibaba/qwen3.8-flash"]
cheapest_list_usd_per_m: "0.15 in, 0.47 out, at alibaba on openrouter"
cheapest_effective_usd_per_m: "0.0298 in, 0.47 out, at alibaba on vercel"
context: 1000000
max_output: 131072
input: ["text", "image", "video"]
output: ["text"]
released: "2026-08-26"
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "include_reasoning", "logit_bias", "logprobs", "max_tokens", "min_p", "presence_penalty", "reasoning", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 5
gateways: "openrouter (3), requesty (1), vercel (1)"
no_training: 5
trains: 0
retains_prompts: 2
---
# Qwen: Qwen3.8 Flash

Qwen3.8-Flash is Qwen’s fast, cost-efficient multimodal model, combining advanced reasoning and generation with a native 1M-token context window. Built for coding, agentic workflows, and visual understanding, it handles large codebases, long documents, charts, videos, and desktop applications.