metadata

base_model: microsoft/Phi-3-medium-4k-instruct
license: mit
license_link: https://huggingface.co/microsoft/Phi-3-medium-4k-instruct/resolve/main/LICENSE
language:
  - multilingual
pipeline_tag: text-generation
model_creator: Microsoft
model_name: Phi 3 mini 4k instruct
model_type: phi-msft
quantized_by: Second State Inc.
tags:
  - nlp
  - code

Phi-3-medium-4k-instruct-GGUF

Original Model

microsoft/Phi-3-medium-4k-instruct

Run with LlamaEdge

LlamaEdge version: coming soon

Prompt template

Prompt type: phi-3-chat

Prompt string

<|system|>
{system_message}<|end|>
<|user|>
{user_message_1}<|end|>
<|assistant|>
{assistant_message_1}<|end|>
<|user|>
{user_message_2}<|end|>
<|assistant|>

Context size: 5120

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
Phi-3-medium-4k-instruct-Q2_K.gguf	Q2_K	2	5.14 GB	smallest, significant quality loss - not recommended for most purposes
Phi-3-medium-4k-instruct-Q3_K_L.gguf	Q3_K_L	3	7.49 GB	small, substantial quality loss
Phi-3-medium-4k-instruct-Q3_K_M.gguf	Q3_K_M	3	6.92 GB	very small, high quality loss
Phi-3-medium-4k-instruct-Q3_K_S.gguf	Q3_K_S	3	6.06 GB	very small, high quality loss
Phi-3-medium-4k-instruct-Q4_0.gguf	Q4_0	4	7.9 GB	legacy; small, very high quality loss - prefer using Q3_K_M
Phi-3-medium-4k-instruct-Q4_K_M.gguf	Q4_K_M	4	8.57 GB	medium, balanced quality - recommended
Phi-3-medium-4k-instruct-Q4_K_S.gguf	Q4_K_S	4	7.95 GB	small, greater quality loss
Phi-3-medium-4k-instruct-Q5_0.gguf	Q5_0	5	9.62 GB	legacy; medium, balanced quality - prefer using Q4_K_M
Phi-3-medium-4k-instruct-Q5_K_M.gguf	Q5_K_M	5	10.1 GB	large, very low quality loss - recommended
Phi-3-medium-4k-instruct-Q5_K_S.gguf	Q5_K_S	5	9.62 GB	large, low quality loss - recommended
Phi-3-medium-4k-instruct-Q6_K.gguf	Q6_K	6	11.5 GB	very large, extremely low quality loss
Phi-3-medium-4k-instruct-Q8_0.gguf	Q8_0	8	14.8 GB	very large, extremely low quality loss - not recommended
Phi-3-medium-4k-instruct-f16.gguf	f16	16	27.9 GB

Quantized with llama.cpp b2961.