Get Started →

Get Started →

qwen3.5-flash

Alibaba Qwen

Qwen3.5 原生视觉语言 Flash 系列基于线性注意力与稀疏混合专家（MoE）的混合架构，在提升推理效率的同时，相较于 Qwen 3 系列，在文本和多模态任务上均实现了性能跃升，并兼顾快速响应速度与整体模型能力。

Try online Integration docs

Pricing

Usage Type	TokenBay Pricing	40% off
Input	-	-
≤128K	¥0.1200/M tokens	¥0.2000/M tokens
≤256K	¥0.4800/M tokens	¥0.8000/M tokens
≤1M	¥0.7200/M tokens	¥1.20/M tokens
Output	-	-
≤128K	¥1.20/M tokens	¥2.00/M tokens
≤256K	¥4.80/M tokens	¥8.00/M tokens
≤1M	¥7.20/M tokens	¥12.00/M tokens
Text Cache Creation	-	-
≤128K	¥0.1500/M tokens	¥0.2500/M tokens
≤256K	¥0.6000/M tokens	¥1.00/M tokens
≤1M	¥0.9000/M tokens	¥1.50/M tokens
Text Cache Read	-	-
≤128K	¥0.0120/M tokens	¥0.0200/M tokens
≤256K	¥0.0480/M tokens	¥0.0800/M tokens
≤1M	¥0.0720/M tokens	¥0.1200/M tokens

Model Info

Base URL

https://api.tokenbay.me

Endpoints

openaiOpenAI Chat Completions

/v1/chat/completions

anthropicAnthropic Messages API

/v1/messages

Protocols

OpenAI-compatibleAnthropic Claude

Tags

-