Model

Qwen3.7 Flash

CLOSED

Alibaba · Qwen3.7 family · released Jul 27, 2026

Cheapest tier of the Qwen3.7 line (model id qwen3.7-flash-20260727), released ~5 weeks after Qwen3.7-Max/Plus. A vision-language reasoning model pitched at multimodal agents, visual coding, search and computer interaction (object recognition, spatial understanding). No independently-verified benchmark suite found at time of writing; treat as unranked until third-party coverage lands.

ReasoningCodingVisionFunction callingTool useAgentic

—

Elo · unrated

Parameters

Undisclosed

Active params

Undisclosed

Context

1M tokens

Architecture

MoE, exact expert configuration undisclosed; cheaper vision-language tier below Qwen3.7-Plus/Max, tuned for multimodal agents, visual coding and computer interaction

License

Proprietary

Languages

—

API price (in/out)

$0.03 / $0.13

Modalities

text · vision · video

Benchmark results

Bar shows position within the tracked field; marker = field best

Run it locally

Closed weights — available via API only. No local deployment.

Input / M tok

$0.03

Output / M tok

$0.13

Qwen3.7 family

Elo progression across releases

Qwen3.7-MaxMay 20262543.8 Qwen3.7-PlusJun 20262422.6 Qwen3.7 FlashJul 20261000.0

Resources

DOCShttps://www.alibabacloud.com/help/en/model-studio/models APIAvailable via Alibaba API

API price $0.03/$0.13 · each benchmark row carries its own source badge (see methodology)