Yes. Start with free quota in iMini Agent; continue on a subscription when quota runs out. No separate Qwen API key required.

Qwen · Qwen3.7 Plus · May 2026

Qwen3.7 Plus — multimodal general model balancing capability and efficiency

Q: Is the name on this page the same model in Agent?

Yes. Choosing Qwen3.7 Plus in Agent is the model described here.

Q: What are the main specs?

1M-token context, 65K max output, text and image input, hybrid thinking on/off.

Q: Does it still support video?

3.7 Plus focuses on text + image. For video input, keep using Qwen3.6 Plus.

Q: How do I choose vs Qwen3.7 Max and Qwen3.6 Plus?

For everyday multimodal, prefer 3.7 Plus. For peak pure text, use Max. When video is required, use 3.6 Plus.

Q: Can outputs be used commercially?

Yes. Content you create on iMini carries commercial use rights under the product terms.

Million-token context
Text + image input
Hybrid thinking on/off
Built-in tools ready

What is Qwen3.7 Plus?

Qwen3.7 Plus is the balanced tier of Tongyi Qianwen’s Qwen3.7 series—performance, cost, and tool use—with text and image input. Versus Qwen3.6 Plus, context and tooling keep improving; for peak pure-text reasoning, use Qwen3.7 Max. Pick Qwen3.7 Plus in iMini Agent to use it.

Vendor: Qwen
Released: May 2026
Context: 1M tokens
Max output: 65K tokens
Deep thinking: Hybrid thinking · on/off
Best as: Everyday Qwen multimodal workhorse

What’s new vs Qwen3.6 Plus

Balanced pick, vision, and tools—the changes to check before you make Plus your workhorse.

A strong everyday starting point

Officially recommended as a common start for coding and chat; raise to Max for peak jobs.

Text + image input

Screenshots and document images join the same job—fit for UI coding and material synthesis.

Million-token context

Long specs and multi-turn history can stay in one thread.

Built-in tools and hybrid thinking

Tool calling and on/off thinking plug into Agent pipelines.

Official benchmarks

Official charts from the Qwen team’s “Qwen3.7-Plus: Multimodal Agent Intelligence.”

Qwen3.7-Plus multimodal and Agent aggregate scoreboard — Aggregate scoreboard for multimodal Agents: vision understanding, tool use, and Agent tasks on one board. Places Plus vs Max—more audiovisual and multimodal workflow, not pure-text peak.

Three common workflows

Qwen3.7 Plus fits everyday workhorse use on the Qwen stack.

Everyday coding

Feature work and UI changes with screenshots.

Chat and office drafts

Q&A, explainers, and formatted output.

Short-to-mid Agents with tools

Function calling and built-in tool flows. Raise to Max for peak reasoning.

Vs Qwen3.7 Max and Qwen3.6 Plus

Same Qwen family. Gaps are peak capacity, generation, and video modality.

Dimension	Qwen3.7 Plus	Qwen3.7 Max	Qwen3.6 Plus
Context	1M tokens	1M tokens	1M tokens
Reasoning	Hybrid thinking · balanced	Hybrid thinking · peak	Agent coding focus
Coding / Agents	Everyday coding workhorse	Peak text flagship	Multimodal Agents
Long tool runs	Built-in tools + images	Tools & structured output	Built-in tools + video
Speed posture	Performance/cost more balanced	Peak quality first	3.6 balanced
Prefer when	New everyday Qwen multimodal	Qwen peak text reasoning	Still need video input

Using Qwen3.7 Plus on iMini

Free quota, model identity, key specs, and how it differs from Max and 3.6.

Can I try it free?

Is the name on this page the same model in Agent?

What are the main specs?

Does it still support video?

How do I choose vs Qwen3.7 Max and Qwen3.6 Plus?

Can outputs be used commercially?

Use Qwen3.7 Plus free on iMini Agent

Million-token context with text + image—for everyday coding and chat.

Open Qwen3.7 Plus

No Qwen API key required.