[Enhancement]: LibreChat support for gab.ai API

[OrcVole]

What features would you like to see added?

Is your feature request related to a problem? Please describe.

Yes, LibreChat provides excellent support for APIs from providers like Anthropic (Claude), Google (Gemini), and OpenAI, allowing users to seamlessly switch between models. However, it currently lacks integration with Gab AI (https://gab.ai), which offers its own API with unique models emphasizing free speech and alternative perspectives. This limits users who want to access Gab's models directly within LibreChat, often forcing workarounds like routing through tools such as OpenWebUI. Adding this would enhance LibreChat's versatility as a multi-provider platform, making more AI options available without external dependencies.

Describe the solution you'd like

I'd like to see native support for Gab AI's API, implemented similarly to existing providers (e.g., via a new file in /api/providers/ like gab.js, mirroring anthropic.js or google.js). Key features could include:

• API Configuration: Users input their Gab AI API key in the settings panel and select from available models (e.g., "gab-ai") in the model dropdown.
• OpenAI Compatibility: Gab's API is compatible with OpenAI libraries, using baseURL: 'https://gab.ai/v1' and standard endpoints like /chat/completions. This should make integration straightforward, with minimal custom parsing needed.
• Core Functionality: Support for chat completions, conversation history, rate limiting, and error handling, ensuring parity with other providers.
• UI Integration: Add Gab as a selectable provider in the interface, with optional model-specific tweaks if needed (e.g., handling any unique response formats).

This would allow users to access additional models easily, expanding LibreChat's ecosystem while aligning with its open-source, user-centric design.

For reference, here's a quick example from Gab AI's documentation showing OpenAI-compatible usage:

Node.js Example:

const OpenAI = require('openai');

const client = new OpenAI({
  apiKey: 'YOUR_API_KEY_HERE',
  baseURL: 'https://gab.ai/v1',
});

const response = await client.chat.completions.create({
  model: 'gab-ai',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Hello, how are you?' }
  ]
});

console.log(response.choices[0].message.content);

Python Example:

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY_HERE",
    base_url="https://gab.ai/v1",
)

response = client.chat.completions.create(
    model="gab-ai",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello, how are you?"}
    ]
)
print(response.choices[0].message.content)

Describe alternatives you've considered

• Custom Proxy or Middleware: Users could set up a proxy to route Gab API calls through an existing OpenAI-compatible endpoint, but this adds complexity and isn't as seamless as native integration.
• External Tools: Routing through apps like OpenWebUI works but fragments the experience, requiring users to manage multiple interfaces instead of centralizing everything in LibreChat.
• Waiting for Gab's Own Tools: Gab might expand its chat interface, but integrating directly into LibreChat would provide a unified, open-source solution for multi-provider use.

Native support in LibreChat feels like the best fit, given its existing architecture for handling similar APIs.

Additional context

Gab AI's API details can be found via their developer resources (e.g., at https://gab.ai – note: API full-access requires a Gab account and key generation). It currently supports models like "Qwen 3 235b", "Arya", "Moonshot Kimi K2" and potentially others, but with very reasonable pricing and rate limits similar to other providers (e.g., token-based). This integration could attract users from communities valuing less censored AI, boosting LibreChat's adoption.

I'm happy to provide more details, test implementations, or even contribute a draft pull request if helpful – I've reviewed the repo and believe this could build directly on the OpenAI provider code. Thanks for maintaining such an awesome project!

More details

https://gab.ai

Which components are impacted by your request?

No response

Pictures

No response

Code of Conduct

• I agree to follow this project's Code of Conduct

[OrcVole]

✅ Confirmation & small tweaks for Gab AI integration

I’ve just tested the current Gab AI endpoints; everything is 100 % OpenAI-compatible except for two tiny details that will save you some 400-errors:

1. Base URL
   https://gab.ai/v1 (note the trailing /v1, not /v1/).

2. Model names
   You can hard-code the public list or hit GET https://gab.ai/v1/models for the live catalog.
   Current stable aliases:

   • gab-ai (points to the default Arya 2.0 model)
   • qwen-3-235b
   • moonshot-kimi-k2

3. Headers
   Standard Authorization: Bearer <key> is sufficient; no extra org-id or project-id required.

That’s it—no custom parsing, no streaming quirks. I can open a PR that adds gab.js mirroring the openai.js provider if the maintainers are happy with the above. Just give the 👍 and I’ll push it.

API docs: https://gab.ai/docs (requires login → “API Keys” tab).

[OrcVole]

Adding the Gab AI provider is still desired.
I can supply a clean PR identical to the OpenAI provider (single provider file + librechat.yaml entry + tests).
If maintainers give the 👍, I’ll push and we can merge immediately.
Let me know ☺️