Qwen 3.7 Max

Alibaba (Qwen)

Apache-2.0 hosted frontier model. Top multilingual and agentic-coding open option.

FrontierApache 2.01M contextReleased Mar 2026textvisioncode

API price · per 1M tokens

$2.5/$7.5

input / output

≈ $3.75/M blended

Benchmarks

Bar shows this model's score; tick shows the best score across all tracked models for context.

SWE-bench VerifiedReal GitHub issues, human-verified patches. The standard coding-agent benchmark.

81.0/ best 90.1

SWE-bench Pro2,294 harder real-world GitHub issues. Where the frontier separates now.

60.6/ best 69.2

Terminal-Bench 2.xAgentic command-line workflows — planning, iteration, tool coordination.

69.7/ best 84.0

GPQA DiamondGoogle-proof graduate-level science Q&A. General reasoning proxy.

88.4/ best 94.4

MMLU-ProMulti-task language understanding across 120+ academic subjects.

84.0/ best 93.0

HumanEval+Functional code generation from docstrings.

92.0/ best 96.0

MATH-500Competition-style math problems.

95.0/ best 97.8

Humanity's Last ExamHardest reasoning benchmark (with tools). ~50% = near frontier.

46.0/ best 57.9

AIME 2025American Invitational Mathematics Examination.

90.0/ best 96.5

Other Alibaba (Qwen) models

Qwen 3.5 Coder

Open-weight

≈$0.30/M

All models