metadata
base_model: microsoft/phi-2
inference: false
language:
- en
license: other
license_link: https://huggingface.co/microsoft/phi-3/resolve/main/LICENSE
license_name: microsoft-research-license
model_creator: Microsoft
model_name: Phi 3
model_type: phi-msft
pipeline_tag: text-generation
prompt_template: |
Instruct: {prompt}
Output:
quantized_by: HDKLK
tags:
- nlp
- code
Provided files
| Name | Quant method | Bits | Size | Max RAM required | Use case |
|---|---|---|---|---|---|
| Phi-3-mini-128k-instructQ2_K.gguf | Q2_K | 2 | 1.42 GB | 3.67 GB | smallest, significant quality loss - not recommended for most purposes |
| Phi-3-mini-128k-instructQ3_K_L.gguf | Q3_K_L | 3 | 2.09 GB | 4.10 GB | small, substantial quality loss |
| Phi-3-mini-128k-instructQ3_K_M.gguf | Q3_K_M | 3 | 1.96 GB | 3.98 GB | very small, high quality loss |
| Phi-3-mini-128k-instructQ3_K_S.gguf | Q3_K_S | 3 | 1.68 GB | 3.75 GB | very small, high quality loss |
| Phi-3-mini-128k-instructQ4_K_M.gguf | Q4_K_M | 4 | 2.39 GB | 4.29 GB | medium, balanced quality - recommended |
| Phi-3-mini-128k-instructQ4_K_S.gguf | Q4_K_S | 4 | 2.19 GB | 4.12 GB | small, greater quality loss |
| Phi-3-mini-128k-instructQ5_K_M.gguf | Q5_K_M | 5 | 2.82 GB | 4.57 GB | large, very low quality loss - recommended |
| Phi-3-mini-128k-instructQ5_K_S.gguf | Q5_K_S | 5 | 2.64 GB | 4.43 GB | large, low quality loss - recommended |
| Phi-3-mini-128k-instructQ6_K.gguf | Q6_K | 6 | 3.14 GB | 4.79 GB | very large, extremely low quality loss |
| Phi-3-mini-128k-instructQ8_0.gguf | Q8_0 | 8 | 4.06 GB | 5.46 GB | very large, extremely low quality loss - not recommended |
How to download GGUF files
The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
- LM Studio
- LoLLMS Web UI
- Faraday.dev