Hugging Face's logo

RichardErkhov
/
frankenmerger_-_cosmo-3b-test-gguf

GGUF
Inference Endpoints
Model card Files Community
RichardErkhov commited on
Commit
1ed67a1
•
1 Parent(s): 0598d59

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +91 -0
README.md ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Quantization made by Richard Erkhov.
2
+
3
+ [Github](https://github.com/RichardErkhov)
4
+
5
+ [Discord](https://discord.gg/pvy7H8DZMG)
6
+
7
+ [Request more models](https://github.com/RichardErkhov/quant_request)
8
+
9
+
10
+ cosmo-3b-test - GGUF
11
+ - Model creator: https://huggingface.co/frankenmerger/
12
+ - Original model: https://huggingface.co/frankenmerger/cosmo-3b-test/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [cosmo-3b-test.Q2_K.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q2_K.gguf) | Q2_K | 1.03GB |
18
+ | [cosmo-3b-test.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.IQ3_XS.gguf) | IQ3_XS | 1.14GB |
19
+ | [cosmo-3b-test.IQ3_S.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.IQ3_S.gguf) | IQ3_S | 1.21GB |
20
+ | [cosmo-3b-test.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q3_K_S.gguf) | Q3_K_S | 1.21GB |
21
+ | [cosmo-3b-test.IQ3_M.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.IQ3_M.gguf) | IQ3_M | 1.26GB |
22
+ | [cosmo-3b-test.Q3_K.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q3_K.gguf) | Q3_K | 1.34GB |
23
+ | [cosmo-3b-test.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q3_K_M.gguf) | Q3_K_M | 1.34GB |
24
+ | [cosmo-3b-test.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q3_K_L.gguf) | Q3_K_L | 1.46GB |
25
+ | [cosmo-3b-test.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.IQ4_XS.gguf) | IQ4_XS | 1.49GB |
26
+ | [cosmo-3b-test.Q4_0.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q4_0.gguf) | Q4_0 | 1.56GB |
27
+ | [cosmo-3b-test.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.IQ4_NL.gguf) | IQ4_NL | 1.57GB |
28
+ | [cosmo-3b-test.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q4_K_S.gguf) | Q4_K_S | 1.57GB |
29
+ | [cosmo-3b-test.Q4_K.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q4_K.gguf) | Q4_K | 1.67GB |
30
+ | [cosmo-3b-test.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q4_K_M.gguf) | Q4_K_M | 1.67GB |
31
+ | [cosmo-3b-test.Q4_1.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q4_1.gguf) | Q4_1 | 1.73GB |
32
+ | [cosmo-3b-test.Q5_0.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q5_0.gguf) | Q5_0 | 1.9GB |
33
+ | [cosmo-3b-test.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q5_K_S.gguf) | Q5_K_S | 1.9GB |
34
+ | [cosmo-3b-test.Q5_K.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q5_K.gguf) | Q5_K | 1.95GB |
35
+ | [cosmo-3b-test.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q5_K_M.gguf) | Q5_K_M | 1.95GB |
36
+ | [cosmo-3b-test.Q5_1.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q5_1.gguf) | Q5_1 | 2.07GB |
37
+ | [cosmo-3b-test.Q6_K.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q6_K.gguf) | Q6_K | 2.25GB |
38
+ | [cosmo-3b-test.Q8_0.gguf](https://huggingface.co/RichardErkhov/frankenmerger_-_cosmo-3b-test-gguf/blob/main/cosmo-3b-test.Q8_0.gguf) | Q8_0 | 2.92GB |
39
+
40
+
41
+
42
+
43
+ Original model description:
44
+ ---
45
+ widget:
46
+ - text: 'Artificial Intelligence is'
47
+ example_title: Textbook
48
+ group: Completion
49
+ - text: '<s> [INST] How to take care of exotic cars? [/INST] '
50
+ example_title: Wikihow
51
+ group: Completion
52
+ - text: '<s> [INST] Generate a story about a Dark Knight [/INST] '
53
+ example_title: Story
54
+ group: Completion
55
+ inference:
56
+ parameters:
57
+ temperature: 0.6
58
+ top_p: 0.9
59
+ top_k: 30
60
+ repetition_penalty: 1.2
61
+ license: apache-2.0
62
+ language:
63
+ - en
64
+ pipeline_tag: text-generation
65
+ ---
66
+
67
+ ## 💻 Usage
68
+
69
+ ```python
70
+ !pip install -qU transformers accelerate
71
+
72
+ from transformers import AutoTokenizer
73
+ import transformers
74
+ import torch
75
+
76
+ model = "gmonsoon/frankencosmo-test"
77
+ messages = [{"role": "user", "content": "What is a large language model?"}]
78
+
79
+ tokenizer = AutoTokenizer.from_pretrained(model)
80
+ prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
81
+ pipeline = transformers.pipeline(
82
+ "text-generation",
83
+ model=model,
84
+ torch_dtype=torch.float16,
85
+ device_map="auto",
86
+ )
87
+
88
+ outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
89
+ print(outputs[0]["generated_text"])
90
+ ```
91
+