Skip to content
This repository has been archived by the owner on Jun 12, 2024. It is now read-only.

Commit

Permalink
doc: make readme shorter
Browse files Browse the repository at this point in the history
  • Loading branch information
@dsdanielpark
dsdanielpark committed Apr 17, 2024
1 parent f7a7f17 commit 277d827
Showing 1 changed file with 44 additions and 49 deletions.
93 changes: 44 additions & 49 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -135,7 +135,7 @@ pip install -q -U python-gemini-api


> [!IMPORTANT]
> Try different Google accounts until you find a working cookie. Use a fresh browser to ensure no remaining cookie values. Use secret browsing mode with independent cookies. Results may vary depending on factors like IP and account status. Providing the entire set of cookies seems to fix one cookie per account. Additionally, once successfully connected with that cookie, it seems to work flawlessly for over three weeks without any errors. *Try various methods until you succeed. Experiment in different environments.*
> Experiment with different Google accounts and browser settings to find a working cookie. Success may vary by IP and account status. Once connected, a cookie typically remains effective over a month. Keep testing until successful.

<br>
Expand Down Expand Up @@ -211,8 +211,8 @@ GeminiClient = Gemini(cookies=cookies)
# GeminiClient = Gemini(auto_cookies=True) # Or use auto_cookies paprameter
```

#### Auto Cookie Update
For `auto_cookie` to be set to `True`, Gemini WebUI must be active in the browser. The [browser_cookie3](https://github.com/borisbabic/browser_cookie3) enables automatic cookie collection, though updates may not be complete yet.
##### Auto Cookie Update
For `auto_cookie` to be set to `True`, and adjust `target_cookies`. Gemini WebUI must be active in the browser. The [browser_cookie3](https://github.com/borisbabic/browser_cookie3) enables automatic cookie collection, though updates may not be complete yet.



Expand All @@ -234,9 +234,7 @@ print(response.payload)
<br>

The output of the generate_content function is `GeminiModelOutput`, with the following structure:

**Properties of GeminiModelOutput:**
The output of the generate_content function is `GeminiModelOutput`, with the following structure:
- *rcid*: returns the response candidate id of the chosen candidate.
- *text*: returns the text of the chosen candidate.
- *code*: returns the codes of the chosen candidate.
Expand All @@ -263,8 +261,7 @@ response_text, response_status = GeminiClient.send_request("Hello, Gemini. What'
print(response_text)
```
You can track the total number of requests made by accessing the `request_count` property within the `Gemini` class.
> [!NOTE]
> Cookies can change quickly. **Don't reopen the same session or repeat prompts too often**; they'll expire faster. If the cookie value doesn't export correctly, refresh the Gemini page and export again.

<br>

### # 04. Text generation
Expand All @@ -281,26 +278,25 @@ print(response.text)
### # 05. Image generation
Returns images generated by Gemini.

*Async downloader*

*Sync downloader*
```python
from gemini import Gemini, GeminiImage

response = GeminiClient.generate_content("Create illustrations of Seoul, South Korea.")
generated_images = response.generated_images # Check generated images [Dict]

GeminiImage.save_sync(generated_images, save_path="save_dir", cookies=cookies)
generated_images = response.generated_images # Check generated images [Dict]

# You can use byte type image dict for printing images as follow:
# bytes_images_dict = GeminiImage.fetch_images_dict_sync(generated_images, cookies=cookies) # Get bytes images dict
# GeminiImage.save_images_sync(bytes_images_dict, path="save_dir", cookies=cookies) # Save to path
await GeminiImage.save(generated_images, "save_dir", cookies=cookies)
# image_data_dict = await GeminiImage.fetch_images_dict(generated_images, cookies=cookies)
# await GeminiImage.save_images(image_data_dict, "save_dir")
```

<details><summary>Display images in IPython</summary>


<details><summary>Further</summary>

*Display images in IPython*
You can display the image or transmit it to another application in byte format.

```
```python
bytes_images_dict = GeminiImage.fetch_images_dict_sync(generated_images, cookies) # Get bytes images dict
from IPython.display import display, Image
import io
Expand All @@ -311,26 +307,24 @@ GeminiImage.save_sync(generated_images, save_path="save_dir", cookies=cookies)
display(image)
```

</details>



*Async downloader*

*Sync downloader*
```python
response = GeminiClient.generate_content("Create illustrations of Seoul, South Korea.")
from gemini import Gemini, GeminiImage

response = GeminiClient.generate_content("Create illustrations of Seoul, South Korea.")
generated_images = response.generated_images # Check generated images [Dict]

await GeminiImage.save(generated_images, "save_dir", cookies=cookies)
# image_data_dict = await GeminiImage.fetch_images_dict(generated_images, cookies=cookies)
# await GeminiImage.save_images(image_data_dict, "save_dir")
```
GeminiImage.save_sync(generated_images, save_path="save_dir", cookies=cookies)

# You can use byte type image dict for printing images as follow:
# bytes_images_dict = GeminiImage.fetch_images_dict_sync(generated_images, cookies=cookies) # Get bytes images dict
# GeminiImage.save_images_sync(bytes_images_dict, path="save_dir", cookies=cookies) # Save to path
```

<details><summary>Async downloader wrapper</summary>
*Async downloader wrapper*

```
```python
import asyncio
from gemini import Gemini, GeminiImage

Expand Down Expand Up @@ -371,19 +365,7 @@ if __name__ == "__main__":
### # 06. Retrieving Images from Gemini Responses
Returns images in response of Gemini.

*Sync downloader*
```python
from gemini import Gemini, GeminiImage

response = GeminiClient.generate_content("Please recommend a travel itinerary for Seoul.")
response_images = response.web_images # Check response images [Dict]

GeminiImage.save_sync(response_images, save_path="save_dir")

# You can use byte type image dict as follow:
# bytes_images_dict = GeminiImage.fetch_bytes_sync(response_images, cookies) # Get bytes images dict
# GeminiImage.save_images_sync(bytes_images_dict, path="save_dir") # Save to path
```
*Async downloader*
```python
response = GeminiClient.generate_content("Create illustrations of Seoul, South Korea.")
Expand All @@ -395,9 +377,25 @@ await GeminiImage.save(response_images, "save_dir")
# await GeminiImage.save_images(image_data_dict, "save_dir")
```

<details><summary>Async downloader wrapper</summary>
<details><summary>Further</summary>


*Sync downloader*
```python
from gemini import Gemini, GeminiImage

response = GeminiClient.generate_content("Please recommend a travel itinerary for Seoul.")
response_images = response.web_images # Check response images [Dict]

GeminiImage.save_sync(response_images, save_path="save_dir")

# You can use byte type image dict as follow:
# bytes_images_dict = GeminiImage.fetch_bytes_sync(response_images, cookies) # Get bytes images dict
# GeminiImage.save_images_sync(bytes_images_dict, path="save_dir") # Save to path
```

*Async downloader wrapper*
```python
import asyncio
from gemini import Gemini, GeminiImage

Expand Down Expand Up @@ -589,7 +587,7 @@ If you have sufficient GPU resources, you can download weights directly instead
### Open-source LLM, [Gemma](https://huggingface.co/google/gemma-7b)


[Gemma](https://huggingface.co/google/gemma-7b) models are Google's lightweight, advanced text-to-text, decoder-only language models, derived from Gemini research. Available in English, they offer open weights and variants, ideal for tasks like question answering and summarization. Their small size enables deployment in resource-limited settings, broadening access to cutting-edge AI. For more infomation, visit [Gemma-7b](https://huggingface.co/google/gemma-7b) model card.
Gemma models are Google's lightweight, advanced text-to-text, decoder-only language models, derived from Gemini research. Available in English, they offer open weights and variants, ideal for tasks like question answering and summarization. For more infomation, visit [Gemma-7b](https://huggingface.co/google/gemma-7b) model card.

#### How to use Gemma
```python
Expand All @@ -607,7 +605,7 @@ print(tokenizer.decode(outputs[0]))

### Open-source LLM, [Code Gemma](https://huggingface.co/collections/google/codegemma-release-66152ac7b683e2667abdee11)

[CodeGemma](https://huggingface.co/blog/codegemma), which is an official release from Google for code LLMs, was released on April 9, 2024. It provides three models specifically designed for generating and interacting with code. You can explore the [Code Gemma models](https://huggingface.co/collections/google/codegemma-release-66152ac7b683e2667abdee11) and view the [model card](https://huggingface.co/google/codegemma-7b-it) for more details.
CodeGemma, which is an official release from Google for code LLMs, was released on April 9, 2024. It provides three models specifically designed for generating and interacting with code. You can explore the [Code Gemma models](https://huggingface.co/collections/google/codegemma-release-66152ac7b683e2667abdee11) and view the [model card](https://huggingface.co/google/codegemma-7b-it) for more details.

#### How to use Code Gemma
```python
Expand All @@ -631,9 +629,6 @@ print(tokenizer.decode(outputs[0]))
## Utilize free open-source LLM API through [Open Router](https://openrouter.ai/)
OpenRouter offers temporary free inference for select models. Obtain an API key from [Open Router API](https://openrouter.ai/keys) and check free models at [Open Router models](https://openrouter.ai/docs#models). Use models with a 0-dollar token cost primarily; other models may incur charges. See more [free open-source LLM API guide](https://github.com/dsdanielpark/Gemini-API/blob/main/documents/README_OPENROUTER.md)

> [!NOTE]
> You can easily receive responses from open LLMs without this package by following the instructions on [here](https://openrouter.ai/docs#models).
```python
from gemini import OpenRouter

Expand Down

0 comments on commit 277d827

Please sign in to comment.