#80Alibaba· released 2026-04-27

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...

View on OpenRouter ↗
Context
262K
Input $/1M
$1.04
Output $/1M
$6.24
Modality
text
Arena Elo
1485
LMArena — human blind votes
AA Index
72.0
Artificial Analysis composite
LiveBench
71.0
Contamination-resistant
GPQA
77.0%
Grad-level science reasoning
MMLU-Pro
79.0%
Broad knowledge
SWE-bench
60.0%
Real GitHub bugs (Verified)
Terminal-Bench
50.0%
Agentic shell tasks
ARC-AGI
19.0%
Abstract reasoning
Aider Polyglot
74.0%
Multi-language code editing
Speed
62 tok/s
Median output speed

Profile

Qwen: Qwen3.6 Max Preview is Alibaba's entry at position #80 on the current aggregated leaderboard. Its strongest signals are real-world software engineering, unattended agentic work. Every figure below is aggregated from the public sources listed on the homepage and refreshed on each pipeline run.

The score spread is comparatively complete for this model, but the standing advice still applies: treat gaps of a few points as ties and validate on your own workload before committing.

How this model's position is computed is documented in the methodology; what each benchmark actually measures is in the benchmark guide. See a figure that disagrees with its source? Report it — accepted corrections apply on the next refresh.

Nearby on the leaderboard