Qwen3.7-Plus
Editorial notes
GA 1 junio 2026 en Alibaba Cloud Model Studio (preview en LM Arena desde ~14 may). Variante multimodal agent de la familia Qwen3.7 (vision + video + GUI/CLI agentic loop), sibling economico de Max (~6x mas barato input). Proprietary/API-only, 1M context, max output 65536. Pricing $0.40/$1.60 per MTok (cache $0.08). AA Intelligence Index 53. SWE-bench-Pro 57.6 corroborado en 3 fuentes. Resto de scores son vendor-stated single-source (qwen.ai blog es SPA ilegible, primary no verificable): GPQA-D 90.3, SWE-V 77.7, OSWorld 73.3, ScreenSpot-Pro 79.0, Terminal-Bench 70.3, MCP-Atlas 76.4, AIME-2025 93.3 (Atlas independiente, thinking on). Marcados vendor-reported hasta que el model card oficial sea legible.
Spec sheet
- Company
- Alibaba
- Country
- CN
- Type
- multimodal
- Release
- 2026-06
- Context
- 1.0M tokens
- License
- proprietary
- Pricing (alibaba)
- $0.4/$1.6/M
- Slug
- qwen3-7-plus
Benchmarks (1)
Cite this model
BibTeX · APA
BibTeX
@misc{frontier-qwen3-7-plus,
title = {Qwen3.7-Plus},
author = {{Alibaba}},
year = {2026},
note = {Frontier Benchmarks AI atlas. Accessed 2026-06-17},
url = {https://frontierbenchmarks.com/models/qwen3-7-plus}
} APA
Alibaba (2026). Qwen3.7-Plus [Large language model]. Frontier Benchmarks AI. Retrieved 2026-06-17, from https://frontierbenchmarks.com/models/qwen3-7-plus
Citation reflects the atlas page, not the original model paper. For the paper, see the "Resources" section above.