Mistral AI

Mistral Small 3.1 24B Base

Multimodal
Zero-eval

by Mistral AI

+
+
+
+
About

Mistral Small 3.1 24B Base represents an updated iteration of the 24B foundation model, developed with architectural refinements and improved training. Built to provide enhanced base capabilities for fine-tuning, it incorporates learnings from previous versions for better downstream performance.

+
+
+
+
Pricing Range
Input (per 1M)$0.10 -$0.10
Output (per 1M)$0.30 -$0.30
Providers1
+
+
+
+
Timeline
AnnouncedMar 17, 2025
ReleasedMar 17, 2025
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Apache 2.0
Performance Overview
Performance metrics and category breakdown

Overall Performance

5 benchmarks
Average Score
62.9%
Best Score
81.0%
High Performers (80%+)
2

Performance Metrics

Max Context Window
256.0K
Avg Throughput
137.1 tok/s
Avg Latency
0ms
+
+
+
+
All Benchmark Results for Mistral Small 3.1 24B Base
Complete list of benchmark scores with detailed information
MMLU
text
0.81
81.0%
Self-reported
TriviaQA
text
0.81
80.5%
Self-reported
MMMU
multimodal
0.59
59.3%
Self-reported
MMLU-Pro
text
0.56
56.0%
Self-reported
GPQA
text
0.38
37.5%
Self-reported
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+