Hugging Face's logo

RichardErkhov
/
zjunlp_-_OceanGPT-14B-v0.1-gguf

GGUF
Inference Endpoints
conversational
Model card Files Community
RichardErkhov commited on
Commit
7f45471
·
verified ·
1 Parent(s): 5c6564d

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +163 -0
README.md ADDED
@@ -0,0 +1,163 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ OceanGPT-14B-v0.1 - GGUF
11
+ - Model creator: https://huggingface.co/zjunlp/
12
+ - Original model: https://huggingface.co/zjunlp/OceanGPT-14B-v0.1/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [OceanGPT-14B-v0.1.Q2_K.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q2_K.gguf) | Q2_K | 5.51GB |
18
+ | [OceanGPT-14B-v0.1.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q3_K_S.gguf) | Q3_K_S | 6.31GB |
19
+ | [OceanGPT-14B-v0.1.Q3_K.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q3_K.gguf) | Q3_K | 6.91GB |
20
+ | [OceanGPT-14B-v0.1.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q3_K_M.gguf) | Q3_K_M | 6.91GB |
21
+ | [OceanGPT-14B-v0.1.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q3_K_L.gguf) | Q3_K_L | 7.3GB |
22
+ | [OceanGPT-14B-v0.1.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.IQ4_XS.gguf) | IQ4_XS | 7.37GB |
23
+ | [OceanGPT-14B-v0.1.Q4_0.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q4_0.gguf) | Q4_0 | 7.62GB |
24
+ | [OceanGPT-14B-v0.1.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.IQ4_NL.gguf) | IQ4_NL | 7.68GB |
25
+ | [OceanGPT-14B-v0.1.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q4_K_S.gguf) | Q4_K_S | 7.98GB |
26
+ | [OceanGPT-14B-v0.1.Q4_K.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q4_K.gguf) | Q4_K | 1.54GB |
27
+ | [OceanGPT-14B-v0.1.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q4_K_M.gguf) | Q4_K_M | 5.08GB |
28
+ | [OceanGPT-14B-v0.1.Q4_1.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q4_1.gguf) | Q4_1 | 1.98GB |
29
+ | [OceanGPT-14B-v0.1.Q5_0.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q5_0.gguf) | Q5_0 | 9.18GB |
30
+ | [OceanGPT-14B-v0.1.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q5_K_S.gguf) | Q5_K_S | 9.34GB |
31
+ | [OceanGPT-14B-v0.1.Q5_K.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q5_K.gguf) | Q5_K | 9.81GB |
32
+ | [OceanGPT-14B-v0.1.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q5_K_M.gguf) | Q5_K_M | 9.81GB |
33
+ | [OceanGPT-14B-v0.1.Q5_1.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q5_1.gguf) | Q5_1 | 9.96GB |
34
+ | [OceanGPT-14B-v0.1.Q6_K.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q6_K.gguf) | Q6_K | 11.46GB |
35
+ | [OceanGPT-14B-v0.1.Q8_0.gguf](https://huggingface.co/RichardErkhov/zjunlp_-_OceanGPT-14B-v0.1-gguf/blob/main/OceanGPT-14B-v0.1.Q8_0.gguf) | Q8_0 | 14.03GB |
36
+
37
+
38
+
39
+
40
+ Original model description:
41
+ ---
42
+ license: mit
43
+ pipeline_tag: text-generation
44
+ tags:
45
+ - ocean
46
+ - text-generation-inference
47
+ - oceangpt
48
+ language:
49
+ - en
50
+ - zh
51
+ datasets:
52
+ - zjunlp/OceanInstruct
53
+ ---
54
+
55
+ <div align="center">
56
+ <img src="logo.jpg" width="300px">
57
+
58
+ **OceanGPT(沧渊): A Large Language Model for Ocean Science Tasks**
59
+
60
+ <p align="center">
61
+ <a href="https://github.com/zjunlp/OceanGPT">Project</a> •
62
+ <a href="https://arxiv.org/abs/2310.02031">Paper</a> •
63
+ <a href="https://huggingface.co/collections/zjunlp/oceangpt-664cc106358fdd9f09aa5157">Models</a> •
64
+ <a href="http://oceangpt.zjukg.cn/">Web</a> •
65
+ <a href="#quickstart">Quickstart</a> •
66
+ <a href="#citation">Citation</a>
67
+ </p>
68
+
69
+
70
+ </div>
71
+
72
+ OceanGPT-14B-v0.1 is based on Qwen1.5-14B and has been trained on a bilingual dataset in the ocean domain, covering both Chinese and English.
73
+
74
+ - ❗**Disclaimer: This project is purely an academic exploration rather than a product. Please be aware that due to the inherent limitations of large language models, there may be issues such as hallucinations.**
75
+
76
+
77
+ ## ⏩Quickstart
78
+ ### Download the model
79
+
80
+ Download the model: [OceanGPT-14B-v0.1](https://huggingface.co/zjunlp/OceanGPT-14B-v0.1)
81
+
82
+ ```shell
83
+ git lfs install
84
+ git clone https://huggingface.co/zjunlp/OceanGPT-14B-v0.1
85
+ ```
86
+ or
87
+ ```
88
+ huggingface-cli download --resume-download zjunlp/OceanGPT-14B-v0.1 --local-dir OceanGPT-14B-v0.1 --local-dir-use-symlinks False
89
+ ```
90
+ ### Inference
91
+
92
+ ```python
93
+ from transformers import AutoModelForCausalLM, AutoTokenizer
94
+ import torch
95
+ device = "cuda" # the device to load the model onto
96
+ path = 'YOUR-MODEL-PATH'
97
+ model = AutoModelForCausalLM.from_pretrained(
98
+ path,
99
+ torch_dtype=torch.bfloat16,
100
+ device_map="auto"
101
+ )
102
+ tokenizer = AutoTokenizer.from_pretrained(path)
103
+
104
+ prompt = "Which is the largest ocean in the world?"
105
+ messages = [
106
+ {"role": "system", "content": "You are a helpful assistant."},
107
+ {"role": "user", "content": prompt}
108
+ ]
109
+ text = tokenizer.apply_chat_template(
110
+ messages,
111
+ tokenize=False,
112
+ add_generation_prompt=True
113
+ )
114
+ model_inputs = tokenizer([text], return_tensors="pt").to(device)
115
+
116
+ generated_ids = model.generate(
117
+ model_inputs.input_ids,
118
+ max_new_tokens=512
119
+ )
120
+ generated_ids = [
121
+ output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
122
+ ]
123
+
124
+ response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
125
+ ```
126
+
127
+ ## 📌Models
128
+
129
+ | Model Name | HuggingFace | WiseModel | ModelScope |
130
+ |-------------------|-----------------------------------------------------------------------------------|----------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------|
131
+ | OceanGPT-14B-v0.1 (based on Qwen) | <a href="https://huggingface.co/zjunlp/OceanGPT-14B-v0.1" target="_blank">14B</a> | <a href="https://wisemodel.cn/models/zjunlp/OceanGPT-14B-v0.1" target="_blank">14B</a> | <a href="https://modelscope.cn/models/ZJUNLP/OceanGPT-14B-v0.1" target="_blank">14B</a> |
132
+ | OceanGPT-7B-v0.2 (based on Qwen) | <a href="https://huggingface.co/zjunlp/OceanGPT-7b-v0.2" target="_blank">7B</a> | <a href="https://wisemodel.cn/models/zjunlp/OceanGPT-7b-v0.2" target="_blank">7B</a> | <a href="https://modelscope.cn/models/ZJUNLP/OceanGPT-7b-v0.2" target="_blank">7B</a> |
133
+ | OceanGPT-2B-v0.1 (based on MiniCPM) | <a href="https://huggingface.co/zjunlp/OceanGPT-2B-v0.1" target="_blank">2B</a> | <a href="https://wisemodel.cn/models/zjunlp/OceanGPT-2b-v0.1" target="_blank">2B</a> | <a href="https://modelscope.cn/models/ZJUNLP/OceanGPT-2B-v0.1" target="_blank">2B</a> |
134
+
135
+ ## 🌻Acknowledgement
136
+
137
+ OceanGPT(沧渊) is trained based on the open-sourced large language models including [Qwen](https://huggingface.co/Qwen), [MiniCPM](https://huggingface.co/collections/openbmb/minicpm-2b-65d48bf958302b9fd25b698f), [LLaMA](https://huggingface.co/meta-llama). Thanks for their great contributions!
138
+
139
+ ## Limitations
140
+
141
+ - The model may have hallucination issues.
142
+
143
+ - We did not optimize the identity and the model may generate identity information similar to that of Qwen/MiniCPM/LLaMA/GPT series models.
144
+
145
+ - The model's output is influenced by prompt tokens, which may result in inconsistent results across multiple attempts.
146
+
147
+ - The model requires the inclusion of specific simulator code instructions for training in order to possess simulated embodied intelligence capabilities (the simulator is subject to copyright restrictions and cannot be made available for now), and its current capabilities are quite limited.
148
+
149
+
150
+ ### 🚩Citation
151
+
152
+ Please cite the following paper if you use OceanGPT in your work.
153
+
154
+ ```bibtex
155
+ @article{bi2023oceangpt,
156
+ title={OceanGPT: A Large Language Model for Ocean Science Tasks},
157
+ author={Bi, Zhen and Zhang, Ningyu and Xue, Yida and Ou, Yixin and Ji, Daxiong and Zheng, Guozhou and Chen, Huajun},
158
+ journal={arXiv preprint arXiv:2310.02031},
159
+ year={2023}
160
+ }
161
+
162
+ ```
163
+