Private AI
Qwen3.6 35B A3B is a native vision-language MoE model with hybrid attention. Compared to Qwen3.5 35B A3B, Alibaba reports stronger agentic coding, mathematical and code reasoning, and better spatial understanding (including object localization and detection).
Added Apr 19, 2026
Model weightsContext Window
262.1K
Max Output
16.4K
Avg output tokens (7d)
1.4K tokens
Input Price (Auto)
$0.12/1M
Output Price (Auto)
$0.84/1M
Cache Read (Auto)
$0.059/1M
Capabilities
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
31.6
Choose explicit providers for this model. Auto routing remains available as the default option.
Loading provider options…
Coding Index
41.9
GPQA Diamond
Graduate-level scientific reasoning
84.1%
Better than 81% of models compared
HLE
Humanity's Last Exam
20.2%
Better than 81% of models compared
IFBench
Instruction-following benchmark
64.4%
Better than 75% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
95.3%
Better than 94% of models compared
AA-LCR
Long context reasoning evaluation
63.7%
Better than 78% of models compared
SciCode
Python programming for scientific computing
35.8%
Better than 57% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
34.8%
Last updated Jul 10, 2026
Artificial AnalysisBetter than 80% of models compared
