Overview
Qwen3 VL Flash is a fast, cost-effective vision-language model in the Qwen3-VL series. It supports 33 languages and accepts text and image inputs with tiered pricing across three context tiers for efficient multimodal understanding.
Key Features
- 33 languages — Broad multilingual support for global use cases
- Vision-language — Accepts text and image inputs for multimodal reasoning
- Ultra-affordable — Starting at $0.05 input / $0.40 output per 1M tokens
- Three pricing tiers — 32K, 128K, and 256K context boundaries
- Fast inference — Optimized for speed-sensitive scenarios with high cost-performance ratio
- OpenAI-compatible API — Drop-in replacement using chat completions format
Pricing
Per Million Tokens
0 < Tokens <= 32K
Input Price: $0.05
Output Price: $0.4
32K < Tokens <= 128K
Input Price: $0.075
Output Price: $0.6
128K < Tokens
Input Price: $0.12
Output Price: $0.96
Parameters
No input schema available.