/ Models

Language models

Qwen: Qwen3.6 Max Preview

Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse mixture-of-experts architecture with approximately 1 trillion total parameters. It is optimized for agentic coding, tool use, and...

Maker
Qwen
Context window
262K
Max output
66K
Input / 1M
$1.03
Output / 1M
$6.16
1M in + 1M out
$7.19

Everything else

Max output66K
Input / 1M$1.03
Output / 1M$6.16
Cache read / 1M
Cache write / 1M$1.28
1M in + 1M out$7.19
Takestext
Returnstext
Toolsyes
Reasoningoptional
Open weightsno
Released2026-04-27
Knowledge cutoff
Retires
Catalog idqwen/qwen3.6-max-preview

Benchmarks

sourceepoch
gpqa0.891
gpqaStderr0.0187
sweBenchVerified0.767
sweBenchVerifiedStderr0.0192
frontierMath0.231
frontierMathStderr0.0248
aime0.911
aimeStderr0.0374
simpleQA0.569
simpleQAStderr0.0157

Benchmark scores by Epoch AI, CC BY 4.0.

Using it

In 00, this model is picked per agent — and per tier, so the model that answers your customers need not be the one that writes your code. Get 00.

← all models