Ministral 8B Instruct
Zero-eval
#1MBPP pass@1
#1French MMLU
by Mistral AI
+
+
+
+
About
Ministral 8B Instruct is a language model developed by Mistral AI. It achieves strong performance with an average score of 63.3% across 11 benchmarks. It excels particularly in MT-Bench (83.0%), Winogrande (75.3%), ARC-C (71.9%). It supports a 256K token context window for handling large documents. The model is available through 1 API provider. Released in 2024, it represents Mistral AI's latest advancement in AI technology.
+
+
+
+
Pricing Range
Input (per 1M)$0.10 -$0.10
Output (per 1M)$0.10 -$0.10
Providers1
+
+
+
+
Timeline
AnnouncedOct 16, 2024
ReleasedOct 16, 2024
+
+
+
+
License & Family
License
Mistral Research License
Performance Overview
Performance metrics and category breakdown
Overall Performance
11 benchmarks
Average Score
63.3%
Best Score
83.0%
High Performers (80%+)
1Performance Metrics
Max Context Window
256.0KAvg Throughput
0.1 tok/sAvg Latency
0ms+
+
+
+
All Benchmark Results for Ministral 8B Instruct
Complete list of benchmark scores with detailed information
| MT-Bench | text | 0.83 | 83.0% | Self-reported | |
| Winogrande | text | 0.75 | 75.3% | Self-reported | |
| ARC-C | text | 0.72 | 71.9% | Self-reported | |
| Arena Hard | text | 0.71 | 70.9% | Self-reported | |
| MBPP pass@1 | text | 0.70 | 70.0% | Self-reported | |
| TriviaQA | text | 0.66 | 65.5% | Self-reported | |
| MMLU | text | 0.65 | 65.0% | Self-reported | |
| French MMLU | text | 0.57 | 57.5% | Self-reported | |
| MATH | text | 0.55 | 54.5% | Self-reported | |
| AGIEval | text | 0.48 | 48.3% | Self-reported |
Showing 1 to 10 of 11 benchmarks