Qwen3.7-Max

Published

May 19, 2026

Qwen3.7-Max is an AI model developed by Alibaba (China), first published in May 2026. It works in the language domain, on tasks such as language modeling/generation and agentic tasks.

Epoch AI has no training-compute estimate for this model.

Access: API access. Its weights are not openly released. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Alibaba
Country of organization: China
Domain: Language
Task: Language modeling/generation, Agentic tasks
Model accessibility: API access
Open weights: No
Epoch confidence: Confident

Benchmark results

01	WebDev Arena	Arena Score	1540.8
02	ALE Bench	Performance	1189.4
03	Epoch Capabilities Index	ECI Score	153.1
04	OTIS Mock AIME 2024-2025	mean_score	0.95
05	GPQA Diamond	mean_score	0.92
06	SWE-bench Verified	mean_score	0.77
07	SimpleBench	Score (AVG@5)	0.7
08	SimpleQA Verified	mean_score	0.59
09	SciCode	Score	0.49
10	Proofbench	Accuracy	0.26
11	Chess Puzzles	mean_score	0.22
12	Critpt	Accuracy	0.13
13	Gbaeval	Overall score	0