name = "Qwen3 Max"
description = "Flagship Qwen3 model for coding agents, complex reasoning, and tool use"
family = "qwen"
release_date = "2025-09-23"
last_updated = "2025-09-23"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2025-04"
open_weights = false

[limit]
context = 262_144
output = 65_536

[modalities]
input = ["text"]
output = ["text"]

[[benchmarks]]
name = "Artificial Analysis Coding Index"
score = 26.4
metric = "index"
source = "https://openrouter.ai/qwen/qwen3-max/benchmarks"
date = "2026-05-30"

[[benchmarks]]
name = "SciCode"
score = 38.3
metric = "percent correct"
source = "https://openrouter.ai/qwen/qwen3-max/benchmarks"
date = "2026-05-30"

[[benchmarks]]
name = "Terminal-Bench Hard"
score = 20.5
metric = "success rate"
source = "https://openrouter.ai/qwen/qwen3-max/benchmarks"
date = "2026-05-30"