# Qwen: Qwen3.7 Flash

Provider: qwen  
Category: llm  
Model ID: `qwen/qwen3.7-flash`

Qwen3.7 Flash is a vision-language reasoning model from Alibaba. It is suited for multimodal agents, visual coding, search, and computer interaction, with strengths in object recognition, spatial understanding, and real-world...

## Specs

- Context length: 1000000 tokens
- Max output: 65536 tokens
- Modalities: text, image, video
- Released: 2026-07-27

## Pricing

- Input: $0.03 per million tokens
- Output: $0.13 per million tokens

## Verdict

Qwen3.7 Flash delivers multimodal reasoning across text, image, and video at aggressive pricing—$0.03/$0.13 per Mtok with a million-token context window. It handles long documents and visual analysis without the cost overhead of frontier models. Trade-off: no public benchmarks yet, so performance relative to GPT-4o or Claude Sonnet is unverified. Reach for this when budget matters more than proven leaderboard scores, especially for multimodal batch jobs.

---
Last verified: 2026-09-04T02:00:56.012Z  
Canonical URL: https://switchy.build/directory/models/qwen3-7-flash