Private AI
Mistral Medium 3.5 is a 128B dense open-weights flagship model for instruction-following, reasoning, coding, long-horizon agentic work, tool use, structured output, and multimodal prompts. It supports a 256k context window and configurable reasoning effort.
Added Apr 29, 2026
Context Window
256.0K
Max Output
32.8K
Input Price (Auto)
$1.50/1M
Output Price (Auto)
$7.50/1M
Cache Read (Auto)
$0.75/1M
Capabilities
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
29.9
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
Coding Index
46.9
GPQA Diamond
Graduate-level scientific reasoning
74.8%
Better than 60% of models compared
HLE
Humanity's Last Exam
12.8%
Better than 69% of models compared
IFBench
Instruction-following benchmark
68.8%
Better than 82% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
94.2%
Better than 92% of models compared
AA-LCR
Long context reasoning evaluation
61.0%
Better than 72% of models compared
SciCode
Python programming for scientific computing
39.6%
Better than 70% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
33.3%
Last updated Aug 4, 2026
Artificial AnalysisBetter than 77% of models compared
