Ministral 8B
A state-of-the-art 8B parameter model from Mistral AI's Ministral family, part of 'les Ministraux'. Features special interleaved sliding-window attention pattern for faster and memory-efficient inference, designed for edge computing and on-device applications.
2024-10-16
8B
Decoder-only Transformer with Interleaved Sliding-Window Attention
Mistral Commercial License
Specifications
- Parameters
- 8B
- Architecture
- Decoder-only Transformer with Interleaved Sliding-Window Attention
- License
- Mistral Commercial License
- Context Window
- 128,000 tokens
- Max Output
- 32,000 tokens
- Type
- text
- Modalities
- text
Benchmark Scores
FACTS Grounding56.1
The FACTS Grounding Leaderboard evaluates LLMs' ability to generate factually accurate long-form res...
Advanced Specifications
- Model Family
- Ministral
- API Access
- Available
- Chat Interface
- Not Available
- Hardware Support
- edge-devicesmobile
Capabilities & Limitations
- Capabilities
- function-callingreasoningedge-computinglow-latency-inferencememory-efficient-inference
- Notable Use Cases
- on-device translationinternet-less smart assistantslocal analyticsautonomous roboticsagentic workflowsinput parsingtask routingmulti-step workflows
- Function Calling Support
- Yes