Private AI
Qwen3-30B-A3B-Instruct-2507 is a MoE causal language model featuring 30.5B total parameters and 3.3B active parameters. Running inside a TEE (Trusted Execution Environment), with provider attestation support.
Context Window
262.0K
Max Output
32.8K
Input Price (Auto)
$0.15/1M
Output Price (Auto)
$0.45/1M
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
18.2
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
GPQA Diamond
Graduate-level scientific reasoning
75.3%
Better than 66% of models compared
HLE
Humanity's Last Exam
10.6%
Better than 68% of models compared
IFBench
Instruction-following benchmark
46.1%
Better than 55% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
33.3%
Better than 42% of models compared
AA-LCR
Long context reasoning evaluation
31.2%
Better than 45% of models compared
SciCode
Python programming for scientific computing
36.0%
Better than 61% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
15.2%
AIME 2025
American Invitational Mathematics Examination 2025
71.7%
Better than 66% of models compared
AIME
American Invitational Mathematics Examination
71.7%
Better than 81% of models compared
MMLU-Pro
Professional and academic subject knowledge
82.8%
Better than 82% of models compared
Last updated Jun 24, 2026
Artificial AnalysisBetter than 53% of models compared
LiveCodeBench
Contamination-free coding benchmark
52.4%
Better than 59% of models compared
Math-500
Diverse mathematical problem solving benchmark
98.0%
Better than 91% of models compared
