## Code:
```python
# %%

import os; os.system('pip install -q requests bs4 lxml pandas')
import requests, bs4, json, pandas as pd

def ollama(prompt):
    return requests.post(
        "https://ollama.yauk.tv/api/generate",
        json={
            "model": "llama3.1",
            "format": "json",
            "prompt": prompt,
            "stream": False,
            "options": {
                "temperature": 0,
                "num_ctx": 32768
            }
        }
    ).json()['response']

# %%

url = 'https://ollama.com/library'
response = requests.get(url)
soup = bs4.BeautifulSoup(response.text, 'lxml')
tags = ''.join(str(tag.prettify()) for tag in soup.find_all('li', class_='flex')[:20])
print(tags)

# %%

result = ollama(f''' Parse the following HTML snippet and extract the information into a JSON format. Output only the JSON data, without any additional text, explanation, or formatting.

HTML to analyze: {tags} ''')

df = pd.DataFrame(list(json.loads(result).values())[0])
df
```

## Output:

|    | name              | description                                                                                                                                                                                                                                       | size                                              | pulls   |   tags | updated      |
|---:|:------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------|:--------|-------:|:-------------|
|  0 | llama3.1          | Llama 3.1 is a new state-of-the-art model from Meta available in 8B, 70B and 405B parameter sizes.                                                                                                                                                | ['8B', '70B', '405B']                             | 5.3M    |     94 | 6 days ago   |
|  1 | gemma2            | Google Gemma 2 is a high-performing and efficient model available in three sizes: 2B, 9B, and 27B.                                                                                                                                                | ['2B', '9B', '27B']                               | 1.2M    |     94 | 6 days ago   |
|  2 | qwen2.5           | Qwen2.5 models are pretrained on Alibaba's latest large-scale dataset, encompassing up to 18 trillion tokens.                                                                                                                                     | ['0.5B', '1.5B', '3B', '7B', '14B', '32B', '72B'] | 289.1K  |    133 | 3 days ago   |
|  3 | phi3.5            | A lightweight AI model with 3.8 billion parameters with performance overtaking similarly and larger sized models.                                                                                                                                 | ['3B']                                            | 53.9K   |     17 | 4 weeks ago  |
|  4 | nemotron-mini     | A commercial-friendly small language model by NVIDIA optimized for roleplay, RAG QA, and function calling.                                                                                                                                        | []                                                | 7,650   |     17 | 2 days ago   |
|  5 | mistral-small     | Mistral Small is a lightweight model designed for cost-effective use in tasks like translation and summarization.                                                                                                                                 | ['22B']                                           | 5,816   |     17 | 4 days ago   |
|  6 | mistral-nemo      | A state-of-the-art 12B model with 128k context length, built by Mistral AI in collaboration with NVIDIA.                                                                                                                                          | ['12B']                                           | 202.2K  |     17 | 4 hours ago  |
|  7 | deepseek-coder-v2 | An open-source Mixture-of-Experts code language model that achieves performance comparable to GPT4-Turbo in code-specific tasks.                                                                                                                  | ['16B', '236B']                                   | 294.4K  |     65 | 3 months ago |
|  8 | mistral           | The 7B model released by Mistral AI, updated to version 0.3.                                                                                                                                                                                      | ['7B']                                            | 3.4M    |     84 | 4 months ago |
|  9 | mixtral           | A set of Mixture of Experts (MoE) model with open weights by Mistral AI in 8x7b and 8x22b parameter sizes.                                                                                                                                        | ['8x7B', '8x22B']                                 | 415.8K  |     69 | 5 months ago |
| 10 | codegemma         | CodeGemma is a collection of powerful, lightweight models that can perform a variety of coding tasks like fill-in-the-middle code completion, code generation, natural language understanding, mathematical reasoning, and instruction following. | ['2B', '7B']                                      | 282.5K  |     85 | 5 months ago |
| 11 | command-r         | Command R is a Large Language Model optimized for conversational interaction and long context tasks.                                                                                                                                              | ['35B']                                           | 203.5K  |     32 | 3 weeks ago  |
| 12 | command-r-plus    | Command R+ is a powerful, scalable large language model purpose-built to excel at real-world enterprise use cases.                                                                                                                                | ['104B']                                          | 95.8K   |     21 | 3 weeks ago  |
| 13 | llava             | 🌋 LLaVA is a novel end-to-end trained large multimodal model that combines a vision encoder and Vicuna for general-purpose visual and language understanding.                                                                                    | ['7B', '13B', '34B']                              | 1.1M    |     98 | 7 months ago |
| 14 | llama3            | Meta Llama 3: The most capable openly available LLM to date.                                                                                                                                                                                      | ['8B', '70B']                                     | 6.2M    |     68 | 4 months ago |
| 15 | gemma             | Gemma is a family of lightweight, state-of-the-art open models built by Google DeepMind.                                                                                                                                                          | ['2B', '7B']                                      | 4.1M    |    102 | 5 months ago |
| 16 | qwen              | Qwen 1.5 is a series of large language models by Alibaba Cloud spanning from 0.5B to 110B parameters.                                                                                                                                             | ['0.5B', '1.8B', '4B', '32B', '72B', '110B']      | 4M      |    379 | 3 months ago |
| 17 | qwen2             | Qwen2 is a new series of large language models from Alibaba group.                                                                                                                                                                                | ['0.5B', '1.5B', '7B', '72B']                     | 3.7M    |     97 | 3 months ago |
| 18 | llama2            | Llama 2 is a collection of foundation language models ranging from 7B to 70B parameters.                                                                                                                                                          | ['7B', '13B', '70B']                              | 2.1M    |    102 | 7 months ago |