Qwen3.7 Max
alibaba/qwen3.7-max
Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks
Capabilities
ReasoningYes
Tool callingYes
Structured outputNo
TemperatureSupported
AttachmentNo
Open weightsNo
Limits
Context1,000,000 tokens
Max inputN/A
Max output65,536 tokens
Details
Familyqwen
Knowledge cutoffN/A
Release date2026-05-21
Last updated2026-05-21
Modalities
Inputtext
Outputtext
Benchmarks
SWE-Bench Verified
80.4
resolved
SWE-Bench Pro
60.6
resolve rate
SWE-Bench Multilingual
78.3
resolve rate
Terminal-Bench
69.7
success rate
GPQA Diamond
92.4
accuracy
Humanity's Last Exam
41.4
accuracy
SciCode
53.5
MCP Atlas
76.4
success rate
NL2Repo
47.2
Community Rating
—
No ratings yet
Sign in to rate
Pricing
Cost per 1M tokens
InputN/A
OutputN/A