[PR #332] docs: document Atlas Cloud as a custom OpenAI-compatible LLM provider #324

New Issue

2026-06-06T22:10:14-04:00

yindo commented

2026-06-06 22:10:14 -04:00

📋 Pull Request Information

Original PR: https://github.com/vxcontrol/pentagi/pull/332
Author: @lucaszhu-hue
Created: 6/4/2026
Status: 🔄 Open

Base: main ← Head: docs/atlascloud-custom-provider

📝 Commits (1)

91f9e0c docs: document Atlas Cloud as a custom OpenAI-compatible LLM provider

📊 Changes

3 files changed (+43 additions, -0 deletions)

View changed files

📝 .env.example (+5 -0)
📝 README.md (+38 -0)
➕ assets/atlas-cloud-logo.png (+0 -0)

📄 Description

Description of the Change

Problem

PentAGI supports many LLM providers, including a generic Custom (LLM_SERVER_*) OpenAI-compatible endpoint, but the docs don't mention Atlas Cloud — a full-modal inference platform that serves the DeepSeek / Qwen / GLM / Kimi / MiniMax models PentAGI already targets, through one OpenAI-compatible API.

Solution

Document Atlas Cloud as a custom OpenAI-compatible provider. No code change — it uses the existing LLM_SERVER_* variables:

README.md: an "Atlas Cloud (OpenAI-compatible)" subsection under Custom LLM Provider Configuration, with a config example, a reasoning-model note, and a collapsed list of all available Atlas chat models.
.env.example: a commented Atlas Cloud example in the Custom LLM provider block.

N/A (no related issue)

Type of Change

📚 Documentation update
🔧 Configuration change

Areas Affected

External Integrations (LLM/Search APIs)
Documentation

Testing and Verification

Test Configuration

PentAGI Version: main (docs/config-only change)
LLM Provider: Atlas Cloud (custom OpenAI-compatible, https://api.atlascloud.ai/v1)

Test Steps

Set LLM_SERVER_URL=https://api.atlascloud.ai/v1, LLM_SERVER_KEY=<key>, LLM_SERVER_MODEL=deepseek-ai/deepseek-v4-pro, LLM_SERVER_PROVIDER=openai.
Send a chat completion through the OpenAI-compatible endpoint.

Test Results

Verified the documented endpoint directly: a non-streaming POST /v1/chat/completions to Atlas with deepseek-ai/deepseek-v4-pro returns HTTP 200 with a valid completion. Docs/config only — no Go/TypeScript paths changed, so go vet / pnpm lint are unaffected.

Documentation Updates

README.md updates
Configuration documentation updates

Checklist

Code Quality

My code follows the project's coding standards (docs only)
I have added/updated necessary documentation

Security

Sensitive information has been properly handled (examples use placeholder keys only)

Compatibility

Changes are backward compatible

Additional Notes

This README lists providers as text (no provider logos), so I kept the Atlas entry text + link to match the existing style. Happy to adjust wording or placement.

🤖 Generated with Claude Code

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/vxcontrol/pentagi/pull/332 **Author:** [@lucaszhu-hue](https://github.com/lucaszhu-hue) **Created:** 6/4/2026 **Status:** 🔄 Open **Base:** `main` ← **Head:** `docs/atlascloud-custom-provider` --- ### 📝 Commits (1) - [`91f9e0c`](https://github.com/vxcontrol/pentagi/commit/91f9e0c9742defca67b6861f49ac61724d444923) docs: document Atlas Cloud as a custom OpenAI-compatible LLM provider ### 📊 Changes **3 files changed** (+43 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `.env.example` (+5 -0) 📝 `README.md` (+38 -0) ➕ `assets/atlas-cloud-logo.png` (+0 -0) </details> ### 📄 Description ### Description of the Change #### Problem PentAGI supports many LLM providers, including a generic Custom (`LLM_SERVER_*`) OpenAI-compatible endpoint, but the docs don't mention Atlas Cloud — a full-modal inference platform that serves the DeepSeek / Qwen / GLM / Kimi / MiniMax models PentAGI already targets, through one OpenAI-compatible API. #### Solution Document Atlas Cloud as a custom OpenAI-compatible provider. **No code change** — it uses the existing `LLM_SERVER_*` variables: - `README.md`: an "Atlas Cloud (OpenAI-compatible)" subsection under *Custom LLM Provider Configuration*, with a config example, a reasoning-model note, and a collapsed list of all available Atlas chat models. - `.env.example`: a commented Atlas Cloud example in the Custom LLM provider block. N/A (no related issue) ### Type of Change - [x] 📚 Documentation update - [x] 🔧 Configuration change ### Areas Affected - [x] External Integrations (LLM/Search APIs) - [x] Documentation ### Testing and Verification #### Test Configuration ```yaml PentAGI Version: main (docs/config-only change) LLM Provider: Atlas Cloud (custom OpenAI-compatible, https://api.atlascloud.ai/v1) ``` #### Test Steps 1. Set `LLM_SERVER_URL=https://api.atlascloud.ai/v1`, `LLM_SERVER_KEY=<key>`, `LLM_SERVER_MODEL=deepseek-ai/deepseek-v4-pro`, `LLM_SERVER_PROVIDER=openai`. 2. Send a chat completion through the OpenAI-compatible endpoint. #### Test Results Verified the documented endpoint directly: a non-streaming `POST /v1/chat/completions` to Atlas with `deepseek-ai/deepseek-v4-pro` returns HTTP 200 with a valid completion. Docs/config only — no Go/TypeScript paths changed, so `go vet` / `pnpm lint` are unaffected. ### Documentation Updates - [x] README.md updates - [x] Configuration documentation updates ### Checklist #### Code Quality - [x] My code follows the project's coding standards (docs only) - [x] I have added/updated necessary documentation #### Security - [x] Sensitive information has been properly handled (examples use placeholder keys only) #### Compatibility - [x] Changes are backward compatible ### Additional Notes This README lists providers as text (no provider logos), so I kept the Atlas entry text + link to match the existing style. Happy to adjust wording or placement. 🤖 Generated with [Claude Code](https://claude.com/claude-code) --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

yindo added the pull-request label 2026-06-06 22:10:14 -04:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: vxcontrol/pentagi#324