feat: Add ModelsLab provider

[adhikjoshi]

feat: Add ModelsLab provider

Summary

Adds ModelsLab as a provider in Continue, giving users access to uncensored Llama 3.1 8B and 70B models with 128K context windows.

What is ModelsLab?

ModelsLab is an AI API platform providing uncensored language models via an OpenAI-compatible endpoint — ideal for code assistance, creative writing, and use cases where standard content filters are too restrictive.

• Website: https://modelslab.com
• Docs: https://docs.modelslab.com/uncensored-chat
• API key: https://modelslab.com/dashboard

Files

New files

• core/llm/llms/ModelsLab.ts — provider implementation
• docs/docs/reference/model-providers/modelslab.md — user documentation

Modified files (one-line patches)

core/llm/llms/index.ts — add import and export:

+import ModelsLab from "./ModelsLab";

 export const ALL_PROVIDERS = {
   ...
+  modelslab: ModelsLab,
   ...
 };

Implementation

Follows the exact same pattern as Together, Groq, Fireworks, and other OpenAI-compatible providers — extends OpenAI base class, sets providerName and defaultOptions.apiBase:

class ModelsLab extends OpenAI {
  static providerName = "modelslab";
  static defaultOptions: Partial<LLMOptions> = {
    apiBase: "https://modelslab.com/uncensored-chat/v1/",
    model: "llama-3.1-8b-uncensored",
  };
}

No new dependencies. No behavior overrides. All streaming, tool calling, and completion logic is inherited from the OpenAI base class.

Usage

{
  "models": [{
    "title": "ModelsLab Llama 3.1 8B",
    "provider": "modelslab",
    "model": "llama-3.1-8b-uncensored",
    "apiKey": "YOUR_MODELSLAB_API_KEY"
  }]
}

Models

Model	Context	Notes
llama-3.1-8b-uncensored (default)	128K	Fast, efficient
llama-3.1-70b-uncensored	128K	Higher quality

Checklist

• Extends OpenAI base class (same pattern as Together, Groq, Fireworks, Novita, etc.)
• static providerName set to "modelslab"
• static defaultOptions with apiBase and model
• Model name alias map (_convertModelName)
• No new dependencies
• Documentation page with usage examples
• Registered in index.ts (see patch above)

---

Continue Tasks: ▶️ 2 queued · ▶️ 7 not started · ✅ 13 no changes — View all

---

Summary by cubic

Adds ModelsLab as a provider to access uncensored Llama 3.1 8B and 70B models with a 128K context window via an OpenAI-compatible endpoint. No behavior changes or new dependencies.

• New Features
  • New provider: modelslab, extends OpenAI base with apiBase https://modelslab.com/uncensored-chat/v1/ and default model llama-3.1-8b-uncensored.
  • Model alias support (llama3.1-8b, llama3.1-70b) mapped to full IDs.
  • Registered in the LLM provider index.
  • Added provider docs under reference/model-providers/modelslab.

^{Written for commit b061917. Summary will update on new commits.}

[github-actions]

All contributors have signed the CLA ✍️ ✅
_{Posted by the CLA Assistant Lite bot.}

[cubic-dev-ai]

No issues found across 3 files

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: b0619178e2

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Wire model alias conversion into request body

This provider defines short-name aliases in MODEL_IDS, but only overrides _convertModelName; the OpenAI request path builds payloads from options.model in _convertArgs and does not invoke _convertModelName. That means configs like "model": "llama3.1-8b" are sent to ModelsLab unchanged and can fail model resolution, even though this file advertises alias support. Apply the conversion before sending requests (for example by overriding _convertArgs).

Useful? React with 👍 / 👎.

[adhikjoshi]

I have read the CLA Document and I hereby sign the CLA