File size: 6,594 Bytes
bde8b55
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "https://ollama.com/library\n"
     ]
    }
   ],
   "source": [
    "from bs4 import BeautifulSoup\n",
    "from requests import get\n",
    "# download the HTML content\n",
    "\n",
    "base_url = 'https://ollama.com'\n",
    "library_url = f'{base_url}/library'\n",
    "print(library_url)\n",
    "html_content = get(library_url).text\n",
    "\n",
    "\n",
    "# Parse the HTML content with BeautifulSoup\n",
    "soup = BeautifulSoup(html_content, 'html.parser')\n",
    "\n",
    "# Extract all the li elements within the ul\n",
    "li_items = soup.select('ul[role=\"list\"] > li')\n",
    "\n",
    "models = []\n",
    "\n",
    "# Iterate over the extracted li elements and print them\n",
    "for li in li_items:\n",
    "    # get first a tag text\n",
    "    sizes = li.div.div.select('span')\n",
    "    sizes = [size.text for size in sizes]\n",
    "\n",
    "    pulls = li.div.select('p')[1].select('span')\n",
    "    # remove svg tags from pulls\n",
    "    pulls = [pull.text[:-1] for pull in pulls]\n",
    "    pulls = pulls[0].split('\\xa0')[0].strip()\n",
    "\n",
    "    model = {\n",
    "        \"name\": li.h2.text.strip(),\n",
    "        \"description\": li.p.text.strip(),\n",
    "        \"url\": f\"{base_url}{li.a['href']}\",\n",
    "        \"params\": sizes,\n",
    "        \"pulls\": pulls      \n",
    "    }\n",
    "    models.append(model)\n",
    "import json\n",
    "with open('models.json', 'w', encoding=\"utf-8\") as file:\n",
    "    file.write(json.dumps(models, indent=4, ensure_ascii=False))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: llama3\n",
      "Model: phi3\n",
      "Model: wizardlm2\n",
      "Model: mistral\n",
      "Model: gemma\n",
      "Model: mixtral\n",
      "Model: llama2\n",
      "Model: codegemma\n",
      "Model: command-r\n",
      "Model: command-r-plus\n",
      "Model: llava\n",
      "Model: dbrx\n",
      "Model: codellama\n",
      "Model: qwen\n",
      "Model: dolphin-mixtral\n",
      "Model: llama2-uncensored\n",
      "Model: deepseek-coder\n",
      "Model: mistral-openorca\n",
      "Model: nomic-embed-text\n",
      "Model: dolphin-mistral\n",
      "Model: phi\n",
      "Model: orca-mini\n",
      "Model: nous-hermes2\n",
      "Model: zephyr\n",
      "Model: llama2-chinese\n",
      "Model: wizard-vicuna-uncensored\n",
      "Model: starcoder2\n",
      "Model: vicuna\n",
      "Model: tinyllama\n",
      "Model: openhermes\n",
      "Model: starcoder\n",
      "Model: openchat\n",
      "Model: dolphin-llama3\n",
      "Model: yi\n",
      "Model: tinydolphin\n",
      "Model: wizardcoder\n",
      "Model: stable-code\n",
      "Model: mxbai-embed-large\n",
      "Model: neural-chat\n",
      "Model: phind-codellama\n",
      "Model: wizard-math\n",
      "Model: starling-lm\n",
      "Model: falcon\n",
      "Model: dolphincoder\n",
      "Model: nous-hermes\n",
      "Model: orca2\n",
      "Model: sqlcoder\n",
      "Model: stablelm2\n",
      "Model: dolphin-phi\n",
      "Model: solar\n",
      "Model: yarn-llama2\n",
      "Model: deepseek-llm\n",
      "Model: codeqwen\n",
      "Model: bakllava\n",
      "Model: all-minilm\n",
      "Model: samantha-mistral\n",
      "Model: llama3-gradient\n",
      "Model: medllama2\n",
      "Model: wizardlm-uncensored\n",
      "Model: xwinlm\n",
      "Model: nous-hermes2-mixtral\n",
      "Model: stable-beluga\n",
      "Model: wizardlm\n",
      "Model: codeup\n",
      "Model: yarn-mistral\n",
      "Model: everythinglm\n",
      "Model: meditron\n",
      "Model: llama-pro\n",
      "Model: magicoder\n",
      "Model: stablelm-zephyr\n",
      "Model: nexusraven\n",
      "Model: codebooga\n",
      "Model: mistrallite\n",
      "Model: llama3-chatqa\n",
      "Model: wizard-vicuna\n",
      "Model: snowflake-arctic-embed\n",
      "Model: llava-llama3\n",
      "Model: goliath\n",
      "Model: open-orca-platypus2\n",
      "Model: moondream\n",
      "Model: duckdb-nsql\n",
      "Model: notux\n",
      "Model: megadolphin\n",
      "Model: notus\n",
      "Model: alfred\n",
      "Model: llava-phi3\n",
      "Model: falcon2\n"
     ]
    }
   ],
   "source": [
    "for model in models:\n",
    "    tagsurl = f\"{model['url']}/tags\"\n",
    "    tags_page = get(tagsurl).text\n",
    "    # Parse the HTML content with BeautifulSoup\n",
    "    soup = BeautifulSoup(tags_page, 'html.parser')\n",
    "    # select links with the class group\n",
    "    tags = soup.select('a.group')\n",
    "    print(f\"Model: {model['name']}\")\n",
    "    model_tags = []\n",
    "    for tag in tags:\n",
    "        # get the parent div of the tag\n",
    "        parent = tag.parent\n",
    "        sizes = parent.parent.select('div.items-baseline')[0].text.strip().split(' • ',2)\n",
    "        # strip each size\n",
    "        sizes = [size.strip() for size in sizes]\n",
    "        model_tags.append({\n",
    "            \"name\": tag.text.strip(),\n",
    "            \"url\": f\"{base_url}{tag['href']}\",\n",
    "            \"size\": sizes[1],\n",
    "            \"hash\": sizes[0],\n",
    "            \"updated\": sizes[2],\n",
    "        })\n",
    "        link = tag['href']\n",
    "        #print(sizes,\"----\")\n",
    "        # get the next sibling of the parent div\n",
    "        sibling = parent.select('span')\n",
    "        if len(sibling) == 1:\n",
    "            hash = sibling[0].text.strip()\n",
    "            if len(sibling) == 3:\n",
    "                size = sibling[2].strip()\n",
    "            else:\n",
    "                pass\n",
    "                #print(sibling)\n",
    "    model[\"tags\"] = model_tags\n",
    "with open('models.json', 'w', encoding=\"utf-8\") as file:\n",
    "    file.write(json.dumps(models, indent=4, ensure_ascii=False))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}