[Enhancement]: Cloudflare AI Workers Integration

[lunar-studio]

What features would you like to see added?

I spent a while trying to get Cloudflare AI Workers to show up on your list of providers within my Portainer YAML but couldn’t manage to do this. If you could enable access to some of their models, this would be fantastic. Thank you.

More details

Cloudflare AI Workers access.

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

[danny-avila]

For cloudflare workers AI, you need to use the openai compatible spec:

Resources:

• https://developers.cloudflare.com/ai-gateway/providers/workersai/#openai-compatible-endpoints
• https://www.librechat.ai/docs/quick_start/custom_endpoints

Based on the LibreChat documentation, here's a YAML configuration file for Cloudflare Workers AI:

version: 1.2.8
cache: true
endpoints:
  custom:
    - name: "Cloudflare Workers AI"
      apiKey: "${CF_API_TOKEN}"
      baseURL: "https://gateway.ai.cloudflare.com/v1/${CF_ACCOUNT_ID}/${CF_GATEWAY_ID}/workers-ai/v1"
      models:
        default: 
          - "@cf/google/gemma-3-12b-it"
          - "@cf/meta/llama-4-scout-17b-16e-instruct"
          - "@cf/qwen/qwq-32b"
          - "@cf/qwen/qwen2.5-coder-32b-instruct"
          - "@cf/deepseek-ai/deepseek-r1-distill-qwen-32b"
          - "@cf/openai/gpt-oss-120b"
        fetch: false
      titleConvo: true
      titleModel: "@cf/google/gemma-3-12b-it"
      modelDisplayLabel: "Cloudflare AI"

Setup Instructions

1. Create the configuration file: Save the above content as librechat.yaml in your LibreChat project root directory.

2. Set environment variables: Add these to your .env file:

CF_API_TOKEN=your_cloudflare_api_token
CF_ACCOUNT_ID=your_account_id
CF_GATEWAY_ID=your_gateway_id

3. Update Docker configuration: If using Docker, create or edit docker-compose.override.yml:

services:
  api:
    volumes:
      - type: bind
        source: ./librechat.yaml
        target: /app/librechat.yaml

Configuration Details

• name: Displays as "Cloudflare Workers AI" in the endpoint selector
• apiKey: References the Cloudflare API token from environment variables
• baseURL: Uses environment variables for account and gateway IDs to construct the full URL
• models/default: Lists all specified models with @cf/openai/gpt-oss-120b as the first (default) option
• fetch: Set to false since Cloudflare Workers AI doesn't support fetching model lists
• titleConvo: Enables automatic conversation title generation
• titleModel: Uses the GPT-OSS model for generating conversation titles
• modelDisplayLabel: Shows "Cloudflare AI" as the provider label

After setting this up and restarting LibreChat, you'll see "Cloudflare Workers AI" as an available endpoint with all the specified models in the dropdown menu.

Note, gpt-oss models did not seem to work due to a spec incompatibility, this seems like something Cloudflare needs to resolve on their end

[lunar-studio]

That was incredibly fast. Thank you Danny you're awesome!