mistralai/Ministral-3-14B-Instruct-2512-BF16
Parameters13.9B
Context256K
Input → output–
Licenseapache-2.0
PrecisionBF16
ArchitectureMistral3ForConditionalGeneration
ReleasedOct 31, 2025
Downloads81.1K
Based on mistralai/Ministral-3-14B-Base-2512
Other sizes and formats in the family
| Model | Params | Precision | Downloads |
|---|---|---|---|
| mistralai/Ministral-3-14B-Instruct-2512-GGUFgguf | 14B | – | 30.9K |
| mistralai/Ministral-3-14B-Instruct-2512fp8 | 13.9B | F8_E4M3 | 118.9K |
| mistralai/Ministral-3-14B-Base-2512 | 13.9B | BF16 | 3.6K |
| mistralai/Ministral-3-8B-Instruct-2512fp8 | 8.9B | F8_E4M3 | 118.6K |
| mistralai/Ministral-3-8B-Base-2512 | 8.9B | BF16 | 8K |
| mistralai/Ministral-3-8B-Instruct-2512-BF16 | 8.9B | BF16 | 35.4K |
| mistralai/Ministral-3-8B-Instruct-2512-GGUFgguf | 8B | – | 43.1K |
| mistralai/Ministral-3-3B-Base-2512 | 4.3B | BF16 | 10.9K |
| mistralai/Ministral-3-3B-Instruct-2512-BF16 | 4.3B | BF16 | 45.4K |
| mistralai/Ministral-3-3B-Instruct-2512fp8 | 3.8B | F8_E4M3 | 102.7K |
| mistralai/Ministral-3-3B-Instruct-2512-GGUFgguf | 3B | – | 44.8K |
| mistralai/Ministral-3-3B-Instruct-2512-ONNXquantized | 3B | – | 583 |
This release in the news
No coverage of this exact release yet.
Recent news about Mistral models
- Oct 17, 2026 · Tutorial / explainerLocal and Distributed Large LLM Inference with MLX and RDMA on Mac Studio
- Oct 17, 2026 · Opinion / analysisllama.cpp has come a long way
- Oct 16, 2026 · Research paperRecognizing Study Lineage Is Not Enough: Over-Crediting Same-Trial Evidence in LLM-Based Medical Verification
- Oct 15, 2026 · Opinion / analysisI benchmarked whether ASR captures a dictated phone number exactly, in German and French. Whisper: 34% exact. Same text written down: 91% [P]
- Oct 14, 2026 · Research paperSpectrumQuant: Spectrum Alignment for LLM Quantization
- Oct 14, 2026 · Research paperBeyond the Trial-and-Error Loop: Hybrid Projection and Automated Tuning for Distributed Training
- Oct 13, 2026 · Research paperSharing AI progress in mathematics
- Oct 13, 2026 · Research paperATTUNE: Agentic Throughput Tuning for LLM Training