Alibaba logo

Qwen3.7 Max

Alibaba · Released Apr 1, 2026

Open Sourcereasoningcodinglong-contextCompare this model

$1.25

Input price / MTok

#15 of 27

$3.75

Output price / MTok

#10 of 27

1M

Context window

#17 of 28

206

Tokens / sec

#2 of 21

77.2

GPQA Diamond

#15 of 19

85.7

AIME 2025

#7 of 10

70.7

LiveCodeBench

#5 of 8

Summary

Qwen3.7 Max is a open-source language model from Alibaba, released on Apr 1, 2026.

At $1.25 input and $3.75 output per million tokens, it ranks #15 of 27 for input price, placing it mid-range on input cost.

Its context window of 1M tokens ranks #17 of 28 for context depth in this dataset.

It targets reasoning, coding, long-context workloads.

Throughput is measured at 206 tokens per second (ranked #2 of 21 for throughput).

Apache 2.0 open weights; available via Alibaba Cloud Model Studio and third-party APIs.

Benchmarks

BenchmarkScoreMax%RankSource
GPQA Diamond77.210077.2%#15/19Alibaba Qwen Team (self-reported in offi...
AIME 202585.710085.7%#7/10Alibaba Qwen Team (self-reported in offi...
LiveCodeBench70.710070.7%#5/8Alibaba Qwen Team (self-reported in tech...

Similar Models

Compare all 6 models

Models with the most shared benchmark coverage and closest scores to Qwen3.7 Max.

Qwen3.7 Max
GoogleGemini 2.5 Pro
xAIGrok 3
GoogleGemini 2.5 Flash
OpenAIo4-mini
AlibabaQwen3 Max
GPQA Diamond#15 of 19
Graduate-level science reasoning
AIME 2025#7 of 10
Advanced math competition problems
LiveCodeBench#5 of 8
Live competitive programming

Compare Qwen3.7 Max side by side

Pre-filled with 5 similar models

Open comparison

Speed source: https://artificialanalysis.ai/models/qwen3-7-max