Skip to content

feat(providers): add CloudGPU - #3324

Merged
esokullu merged 1 commit into
webbrain-one:mainfrom
mrhy7176:cursor/add-cloudgpu-provider-fb5f
Nov 7, 2026
Merged

esokullu merged 1 commit into
webbrain-one:mainfrom
mrhy7176:cursor/add-cloudgpu-provider-fb5f

Conversation

@mrhy7176

@mrhy7176 mrhy7176 commented Nov 7, 2026

Copy link
Copy Markdown
Contributor

Summary

Adds CloudGPU as a built-in provider card in the extended catalog, the same way the Runcrate (#3302), Opper and API Route cards were added. CloudGPU serves DeepSeek, Kimi, GLM, MiniMax, Qwen and more through an OpenAI-compatible API.

  • Card id: cloudgpu, category router, disabled by default, existing OpenAI-compatible adapter.
  • Base URL: https://cloudgpu.app/v1. API keys: cloudgpu.app (the card's apiKeyUrl stays the bare https://cloudgpu.app)
  • Default model: deepseek-v4-flash (131072 tokens, as reported by GET /v1/models). Suggestions: deepseek-v4-pro, glm-5.3, kimi-k3, qwen3.5-397b.
  • Vision is off on the card because the default model is text-only. Tool calls and Ask streaming stay on.
  • No pricing metadata is added. Chrome and Firefox catalogs stay identical.
Files
  • src/{chrome,firefox}/src/providers/provider-catalog.js: one new entry.
  • src/{chrome,firefox}/icons/providers/cloudgpu.svg: CloudGPU logo mark (no scripts, images or external refs).
  • src/{chrome,firefox}/icons/providers/SOURCES.md: icon source row.
  • test/run.js: id list, counts (86 extra cards, 121/120 defaults) and a defaults/suggestions check.
  • test/provider-catalog-integrations.mjs: endpoint, model, flags and icon URL checks for both browsers.
  • docs/providers-and-models.md: id table, counts and a short setup section.
  • README.md: card count 120/119.

Testing

  • npm run test:provider-catalog — passed
  • npm run test:provider-limits — passed
  • WEBBRAIN_TEST_FILTER='extended provider catalog' node test/run.js — 1 passed, 0 failed
  • GET https://cloudgpu.app/v1/models includes deepseek-v4-flash (context_length 131072, text input), deepseek-v4-pro, glm-5.3, kimi-k3, and qwen3.5-397b. No live chat completion was run.

There is no lint script in this repo. CONTRIBUTING.md says there is no CLA.

Disclosure: I work on CloudGPU.

Reviewer notes

  • Catalog position: cloudgpu sits right after clarifai and before cloudferro-sherlock (strictly alphabetical would put it one slot later). In test/run.js it is on its own line in the id list, and in the docs id table it is a new last row, so existing rows are not re-wrapped. Happy to move it.
  • Icon is the CloudGPU mark (three strokes on a dark rounded square), authored by CloudGPU.
Open in Web Open in Cursor 

Add a disabled-by-default CloudGPU card to the extended catalog on Chrome and Firefox, using the existing OpenAI-compatible adapter.

Co-authored-by: CloudGPU <mrhy7176@users.noreply.github.com>
@vercel

vercel Bot commented Nov 7, 2026

Copy link
Copy Markdown

@cursoragent is attempting to deploy a commit to the esokullu's projects Team on Vercel.

A member of the Team first needs to authorize it.

@esokullu
esokullu merged commit 9fe1768 into webbrain-one:main Nov 7, 2026
1 check failed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants