base_model: TencentARC/Mistral_Pro_8B_v0.1
language:
- en
pipeline_tag: text-generation
license: apache-2.0
model_type: mistral
library_name: transformers
inference: false
datasets:
- HuggingFaceTB/cosmopedia
Mistral Pro 8B v0.1
- Model creator: TencentARC
- Original model: Mistral_Pro_8B_v0.1-7b-it
Description
This repo contains GGUF format model files for TencentARC's Mistral Pro 8B v0.1
Original model
- Developed by: TencentARC
Description
Model Description
Mistral-Pro is a progressive version of the original Mistral model, enhanced by the addition of Transformer blocks. It specializes in integrating both general language understanding and domain-specific knowledge, particularly in programming and mathematics.
Development and Training
Developed by Tencent's ARC Lab, Mistral-Pro is an 8 billion parameter model. It's an expansion of Mistral-7B, further trained on code and math corpora.
Intended Use
This model is designed for a wide range of NLP tasks, with a focus on programming, mathematics, and general language tasks. It suits scenarios requiring integration of natural and programming languages.
Performance
Mistral_Pro_8B_v0.1 showcases superior performance on a range of benchmarks. It enhances the code and math performance of Mistral. Furthermore, it matches the performance of the recently dominant model, Gemma.
Overall Performance on Languages, math and code tasks
| Model | ARC | Hellaswag | MMLU | TruthfulQA | Winogrande | GSM8K | HumanEval |
|---|---|---|---|---|---|---|---|
| Gemma-7B | 61.9 | 82.2 | 64.6 | 44.8 | 79.0 | 50.9 | 32.3 |
| Mistral-7B | 60.8 | 83.3 | 62.7 | 42.6 | 78.0 | 39.2 | 28.7 |
| Mistral_Pro_8B_v0.1 | 63.2 | 82.6 | 60.6 | 48.3 | 78.9 | 50.6 | 32.9 |
Limitations
While Mistral-Pro addresses some limitations of previous models in the series, it may still encounter challenges specific to highly specialized domains or tasks.
Ethical Considerations
Users should be aware of potential biases in the model and use it responsibly, considering its impact on various applications.
Quantizon types
| quantization method | bits | size | description | recommended |
|---|---|---|---|---|
| Q2_K | 2 | 3.36 | very small, very high quality loss | ❌ |
| Q3_K_S | 3 | 3.91 GB | very small, high quality loss | ❌ |
| Q3_K_M | 3 | 4.35 GB | small, substantial quality loss | ❌ |
| Q3_K_L | 3 | 4.74 GB | small, substantial quality loss | ❌ |
| Q4_0 | 4 | 5.09 GB | legacy; small, very high quality loss | ❌ |
| Q4_K_S | 4 | 5.13 GB | medium, balanced quality | ✅ |
| Q4_K_M | 4 | 5.42 GB | medium, balanced quality | ✅ |
| Q5_0 | 5 | 6.20 GB | legacy; medium, balanced quality | ❌ |
| Q5_K_S | 5 | 6.20 GB | large, low quality loss | ✅ |
| Q5_K_M | 5 | 6.36 GB | large, very low quality loss | ✅ |
| Q6_K | 6 | 7.37 GB | very large, extremely low quality loss | ❌ |
| Q8_0 | 8 | 9.55 GB | very large, extremely low quality loss | ❌ |
| FP16 | 16 | 18 GB | enormous, negligible quality loss | ❌ |
Usage
You can use this model with the latest builds of LM Studio and llama.cpp.
If you're new to the world of large language models, I recommend starting with LM Studio.