Private AI
Mistral Code Agent Latest is Mistral's direct API alias for devstral-2512, an agentic coding model built for autonomous software engineering, tool use, and long-running code tasks.
Added Jun 2, 2026
Context Window
262.1K
Max Output
32.8K
Input Price (Auto)
$0.40/1M
Output Price (Auto)
$2.00/1M
Capabilities
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
15.5
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
GPQA Diamond
Graduate-level scientific reasoning
59.4%
Better than 40% of models compared
HLE
Humanity's Last Exam
3.6%
Better than 5% of models compared
IFBench
Instruction-following benchmark
38.1%
Better than 33% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
24.9%
Better than 28% of models compared
AA-LCR
Long context reasoning evaluation
30.0%
Better than 42% of models compared
CritPt
Research-level physics reasoning
0.0%
Better than 27% of models compared
SciCode
Python programming for scientific computing
33.1%
Better than 51% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
18.9%
AIME 2025
American Invitational Mathematics Examination 2025
36.7%
Better than 37% of models compared
MMLU-Pro
Professional and academic subject knowledge
76.2%
Better than 53% of models compared
AA-Omniscience Accuracy
Proportion of correctly answered questions
21.0%
Last updated Jun 22, 2026
Artificial AnalysisBetter than 58% of models compared
LiveCodeBench
Contamination-free coding benchmark
44.8%
Better than 51% of models compared
Better than 53% of models compared
AA-Omniscience Hallucination Rate
Rate of incorrect answers among non-correct responses
85.1%
Better than 34% of models compared
