|
# Free AI APIs |
|
|
|
There are APIs offering free access to AI APIs via reverse proxy, and one of the major players, compatible with LibreChat, is ChimeraGPT. |
|
|
|
Feel free to check out the others, but I haven't personally tested them: [Free AI APIs](https://github.com/NovaOSS/free-ai-apis) |
|
|
|
### ChimeraGPT |
|
|
|
Since ChimeraGPT works with LibreChat, and offers Llama2 along with OpenAI models, let's start with that one: [ChimeraGPT](https://discord.gg/chimeragpt) |
|
|
|
> ⚠️ Never trust 3rd parties. Use at your own risk of privacy loss. Your data may be used for AI training at best or for nefarious reasons at worst; this is true in all cases, even with official endpoints: never give an LLM sensitive/identifying information. If something is free, you are the product. If errors arise, they are more likely to be due to the 3rd party, and not this project, as I test the official endpoints first and foremost. |
|
|
|
You will get your API key from the discord server. The instructions are pretty clear when you join so I won't repeat them. |
|
|
|
Once you have the API key, you should adjust your .env file like this: |
|
|
|
```bash |
|
########################## |
|
# OpenAI Endpoint: |
|
########################## |
|
|
|
OPENAI_API_KEY=your-chimera-api-key |
|
# Reverse proxy settings for OpenAI: |
|
OPENAI_REVERSE_PROXY=https://chimeragpt.adventblocks.cc/v1/chat/completions |
|
|
|
# OPENAI_MODELS=gpt-3.5-turbo,gpt-3.5-turbo-16k,gpt-3.5-turbo-0301,text-davinci-003,gpt-4,gpt-4-0314,gpt-4-0613 |
|
``` |
|
|
|
**Note:** The `OPENAI_MODELS` variable is commented out so that the server can fetch chimeragpt/v1/api/models for all available models. Uncomment and adjust if you wish to specify which exact models you want to use. |
|
|
|
It's worth noting that not all models listed by their API will work, with or without this project. The exact URL may also change, just make sure you include `/v1/chat/completions` in the reverse proxy URL if it ever changes. |
|
|
|
You can set `OPENAI_API_KEY=user_provided` if you would like the user to add their own Chimera API key, just be sure you specify the models with `OPENAI_MODELS` in this case since they won't be able to be fetched without an admin set API key. |
|
|
|
## That's it! You're all set. 🎉 |
|
|
|
### Here's me using Llama2 via ChimeraGPT |
|
|
|
 |
|
|
|
### Plugins also work with this reverse proxy (OpenAI models). [More info on plugins here](https://docs.librechat.ai/features/plugins/introduction.html) |
|
 |
|
|