File size: 4,552 Bytes
5e87634
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e2e8a7a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5e87634
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
---
base_model: EpistemeAI/DeepPhi-3.5-mini-instruct
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
- llama-cpp
- gguf-my-repo
license: mit
language:
- en
---

# Triangle104/DeepPhi-3.5-mini-instruct-Q5_K_S-GGUF
This model was converted to GGUF format from [`EpistemeAI/DeepPhi-3.5-mini-instruct`](https://huggingface.co/EpistemeAI/DeepPhi-3.5-mini-instruct) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/EpistemeAI/DeepPhi-3.5-mini-instruct) for more details on the model.

---
Model Summary
-



Reason Phi model for top performing model with it's size of 3.8B.
Phi-3 - synthetic data and filtered publicly available websites - with a
 focus on very high-quality, reasoning dense data. The model belongs to 
the Phi-3 model family and supports 128K token context length. 



	
		
	

Run locally
-




	
		
	

4bit
	



After obtaining the Phi-3.5-mini-instruct model checkpoint, users can use this sample code for inference.


import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline, BitsAndBytesConfig

torch.random.manual_seed(0)

model_path = "EpistemeAI/DeepPhi-3.5-mini-instruct"

# Configure 4-bit quantization using bitsandbytes
quantization_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",  # You can also try "fp4" if desired.
    bnb_4bit_compute_dtype=torch.float16  # Or torch.bfloat16 depending on your hardware.
)

model = AutoModelForCausalLM.from_pretrained(
    model_path,
    device_map="auto",
    torch_dtype=torch.float16,
    trust_remote_code=True,
    quantization_config=quantization_config,
)
tokenizer = AutoTokenizer.from_pretrained(model_path)

messages = [
    {"role": "system", "content": """
You are a helpful AI assistant. Respond in the following format:
<reasoning>
...
</reasoning>
<answer>
...
</answer>"""},
    {"role": "user", "content": "Can you provide ways to eat combinations of bananas and dragonfruits?"},
    {"role": "assistant", "content": "Sure! Here are some ways to eat bananas and dragonfruits together: 1. Banana and dragonfruit smoothie: Blend bananas and dragonfruits together with some milk and honey. 2. Banana and dragonfruit salad: Mix sliced bananas and dragonfruits together with some lemon juice and honey."},
    {"role": "user", "content": "What about solving a 2x + 3 = 7 equation?"},
]

def format_messages(messages):
    prompt = ""
    for msg in messages:
        role = msg["role"].capitalize()
        prompt += f"{role}: {msg['content']}\n"
    return prompt.strip()

prompt = format_messages(messages)

pipe = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
)

generation_args = {
    "max_new_tokens": 500,
    "return_full_text": False,
    "temperature": 0.0,
    "do_sample": False,
}

output = pipe(prompt, **generation_args)
print(output[0]['generated_text'])




	
		
	

Uploaded  model
-	



Developed by: EpistemeAI
License: apache-2.0
Finetuned from model : unsloth/phi-3.5-mini-instruct-bnb-4bit


This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

---
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)

```bash
brew install llama.cpp

```
Invoke the llama.cpp server or the CLI.

### CLI:
```bash
llama-cli --hf-repo Triangle104/DeepPhi-3.5-mini-instruct-Q5_K_S-GGUF --hf-file deepphi-3.5-mini-instruct-q5_k_s.gguf -p "The meaning to life and the universe is"
```

### Server:
```bash
llama-server --hf-repo Triangle104/DeepPhi-3.5-mini-instruct-Q5_K_S-GGUF --hf-file deepphi-3.5-mini-instruct-q5_k_s.gguf -c 2048
```

Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```

Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```

Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/DeepPhi-3.5-mini-instruct-Q5_K_S-GGUF --hf-file deepphi-3.5-mini-instruct-q5_k_s.gguf -p "The meaning to life and the universe is"
```
or 
```
./llama-server --hf-repo Triangle104/DeepPhi-3.5-mini-instruct-Q5_K_S-GGUF --hf-file deepphi-3.5-mini-instruct-q5_k_s.gguf -c 2048
```