# OpenLLM Buddy

> Flat‑rate GPU hosting for Gemma/Qwen/Open Source LLMS. OpenAI‑compatible API. No DevOps. Pay monthly. Deploy in 3 clicks.

OpenLLM Buddy is a managed open-source LLM deployment platform. Customers pick an AI model template, purchase GPU time packs, and receive an OpenAI-compatible API endpoint with API key access. This file helps LLM agents find authoritative pages on the public site.

- Canonical site: https://openllmbuddy.cloud
- Sitemap: https://openllmbuddy.cloud/sitemap.xml
- Robots: https://openllmbuddy.cloud/robots.txt
- Site map: https://openllmbuddy.cloud/site-map

## Product

- [Home](https://openllmbuddy.cloud/): Open AI models, autonomous agents, and GPU self-deploy — the control layer for open AI
- [Agent Buddy](https://openllmbuddy.cloud/agent-buddy): Agent Buddy for Hermes & OpenClaw — all tasks, 10M free tokens, $100 USD/mo unlimited
- [Agent Buddy pricing](https://openllmbuddy.cloud/agent-buddy#pricing): Free: all tasks + 10M tokens; Unlimited: $100 USD/mo all tasks
- [Playground](https://openllmbuddy.cloud/playground): Hourly GPU-backed chat with uncensored models, to try them before subscribing or deploying
- [Self Deploy LLMs](https://openllmbuddy.cloud/selfdeploy): OpenLLM catalog, templates, deploy flow, and pricing
- [Included agent tasks](https://openllmbuddy.cloud/agent-buddy#browse): Full task catalog included in every plan
- [How it works](https://openllmbuddy.cloud/selfdeploy/self-deploy/how-it-works): Deploy flow, time packs, and API access
- [AI Models](https://openllmbuddy.cloud/selfdeploy/self-deploy/models): OpenLLM catalog, GPU class, and flat pack pricing
- [API documentation](https://openllmbuddy.cloud/selfdeploy/self-deploy/api-docs): OpenAI-compatible chat completions reference
- [Integrations](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations): Setup guides for chat UIs, coding tools, SDKs, agents, and automation apps

## Prompts

- [Prompts and use cases](https://openllmbuddy.cloud/prompts): Hermes Agent and OpenClaw prompts, one page each
- [Chief of Staff that routes the team](https://openllmbuddy.cloud/prompts/hermes-chief-of-staff): Hermes Agent — On demand · Hermes workspace team
- [Inbox sorted before you wake up](https://openllmbuddy.cloud/prompts/hermes-inbox-chief): Hermes Agent — Weekday cron · Gmail + Telegram
- [Scored leads with a real first line](https://openllmbuddy.cloud/prompts/hermes-lead-list): Hermes Agent — On demand · search + Google Sheet
- [Overnight demo page for each new lead](https://openllmbuddy.cloud/prompts/hermes-overnight-demos): Hermes Agent — Nightly cron · sheet + VPS
- [One social draft in your voice](https://openllmbuddy.cloud/prompts/hermes-daily-post): Hermes Agent — Daily cron · posts folder + Telegram
- [Weekly list of content gaps](https://openllmbuddy.cloud/prompts/hermes-content-gaps): Hermes Agent — Monday cron · web search
- [Morning briefing before the first tab](https://openllmbuddy.cloud/prompts/hermes-morning-briefing): Hermes Agent — Daily cron · calendar, weather, RSS, Telegram
- [Competitor page watch](https://openllmbuddy.cloud/prompts/hermes-price-monitor): Hermes Agent — Every 12 or 24 hours · fetch + Telegram
- [Research notes that file themselves](https://openllmbuddy.cloud/prompts/hermes-vault-researcher): Hermes Agent — Daily cron · web + markdown vault
- [Route each job to a cheaper model](https://openllmbuddy.cloud/prompts/hermes-cost-router): Hermes Agent — On demand, then weekly · model config
- [Coding jobs through Codex](https://openllmbuddy.cloud/prompts/hermes-codex-jobs): Hermes Agent — On demand · Codex CLI + Telegram
- [Personal CRM from mail and calendar](https://openllmbuddy.cloud/prompts/openclaw-personal-crm): OpenClaw — Daily cron · Gmail + Calendar + SQLite
- [Meeting action items with an approval queue](https://openllmbuddy.cloud/prompts/openclaw-meeting-actions): OpenClaw — Every 5 minutes on workdays · Fathom + Todoist
- [Urgent email, only when it can wait no longer](https://openllmbuddy.cloud/prompts/openclaw-urgent-email): OpenClaw — Every 30 minutes while you are awake · Gmail
- [Knowledge base you fill from a Telegram topic](https://openllmbuddy.cloud/prompts/openclaw-knowledge-base): OpenClaw — On demand from Telegram · SQLite
- [Nightly business council](https://openllmbuddy.cloud/prompts/openclaw-business-council): OpenClaw — Nightly cron · your business data
- [Nightly security review of this codebase](https://openllmbuddy.cloud/prompts/openclaw-security-council): OpenClaw — Nightly at 3:30am · your repo
- [Daily social snapshots](https://openllmbuddy.cloud/prompts/openclaw-social-tracker): OpenClaw — Daily cron · YouTube, Instagram, X, TikTok
- [Video idea pipeline from a Slack mention](https://openllmbuddy.cloud/prompts/openclaw-video-ideas): OpenClaw — Slack mention · Asana + knowledge base
- [Earnings previews that schedule themselves](https://openllmbuddy.cloud/prompts/openclaw-earnings): OpenClaw — Sunday preview, then one-time jobs · Telegram
- [Food and symptom journal](https://openllmbuddy.cloud/prompts/openclaw-food-journal): OpenClaw — Three reminders a day · Telegram + markdown
- [Morning briefing with the room already filled in](https://openllmbuddy.cloud/prompts/openclaw-daily-briefing): OpenClaw — Daily at 7:00am · calendar, CRM, social
- [Telegram topics and Slack, each with one job](https://openllmbuddy.cloud/prompts/openclaw-messaging-setup): OpenClaw — Standing rules · Telegram + Slack
- [Safety rules for a local assistant](https://openllmbuddy.cloud/prompts/openclaw-security-safety): OpenClaw — Standing rules · nightly, weekly, monthly checks
- [Hourly encrypted database backups](https://openllmbuddy.cloud/prompts/openclaw-db-backups): OpenClaw — Hourly · SQLite + Google Drive
- [Hourly git sync that stops on a conflict](https://openllmbuddy.cloud/prompts/openclaw-git-sync): OpenClaw — Hourly · git push
- [A prompting guide for the model you actually run](https://openllmbuddy.cloud/prompts/openclaw-prompt-guide): OpenClaw — On demand, then whenever a prompt file changes
- [Humanizer for anything a person will read](https://openllmbuddy.cloud/prompts/openclaw-humanizer): OpenClaw — Before longer outbound prose
- [Images on demand](https://openllmbuddy.cloud/prompts/openclaw-image-gen): OpenClaw — On demand · Gemini image API + Telegram
- [Short clips on demand](https://openllmbuddy.cloud/prompts/openclaw-video-gen): OpenClaw — On demand · Veo + Telegram
- [Watch a video and pull the useful parts](https://openllmbuddy.cloud/prompts/openclaw-video-watch): OpenClaw — On demand · Gemini video understanding
- [Google Workspace, read-first](https://openllmbuddy.cloud/prompts/openclaw-google-workspace): OpenClaw — Shared connection · Gmail, Calendar, Drive, Docs
- [Platform health council](https://openllmbuddy.cloud/prompts/openclaw-platform-council): OpenClaw — Nightly · repo, logs, databases
- [Newsletter and pipeline, cached locally](https://openllmbuddy.cloud/prompts/openclaw-newsletter-crm): OpenClaw — Daily cron · Beehiiv + HubSpot + SQLite
- [What the assistant spends on models](https://openllmbuddy.cloud/prompts/openclaw-model-costs): OpenClaw — Every model call · JSONL
- [Asana as the system of record for work](https://openllmbuddy.cloud/prompts/openclaw-asana): OpenClaw — On demand and daily sync · Asana
- [Heartbeat that speaks only when something is wrong](https://openllmbuddy.cloud/prompts/openclaw-health-heartbeat): OpenClaw — Daily, weekly, monthly · alerts only
- [Ask before you update OpenClaw](https://openllmbuddy.cloud/prompts/openclaw-self-update): OpenClaw — Nightly at 9:00pm · OpenClaw CLI

## Models

- [Kimi-K3-Uncensored-GGUF](https://openllmbuddy.cloud/selfdeploy/self-deploy/models/kimi-k3-uncensored-gguf): handle `kimi-k3-uncensored-gguf` — Kimi-K3-Uncensored-GGUF is in the OpenLLM catalog on a dedicated 8× H100 80GB. Deploy it with a flat time pack and call your instance throug
- [Qwen-Image-2.1-Uncensored-GGUF](https://openllmbuddy.cloud/selfdeploy/self-deploy/models/qwen-image-2-1-uncensored-gguf): handle `qwen-image-2-1-uncensored-gguf` — Qwen-Image-2.1-Uncensored-GGUF is in the OpenLLM catalog on a dedicated RTX 4090 24GB. Deploy it with a flat time pack and call your instanc
- [LTX-2.5-uncensored-v1.1-FP8](https://openllmbuddy.cloud/selfdeploy/self-deploy/models/ltx-2-5-uncensored-v1-1-fp8): handle `ltx-2-5-uncensored-v1-1-fp8` — LTX-2.5-uncensored-v1.1-FP8 is in the OpenLLM catalog on a dedicated RTX 5090 32GB. Deploy it with a flat time pack and call your instance t
- [gemma-4-31B-it-uncensored](https://openllmbuddy.cloud/selfdeploy/self-deploy/models/gemma-4-31b-it-uncensored): handle `gemma-4-31b-it-uncensored` — gemma-4-31B-it-uncensored is in the OpenLLM catalog on a dedicated RTX 4090 24GB. Deploy it with a flat time pack and call your instance thr
- [GLM-5.3-Flash-Uncensored-FP8](https://openllmbuddy.cloud/selfdeploy/self-deploy/models/glm-5-3-flash-uncensored-fp8): handle `glm-5-3-flash-uncensored-fp8` — GLM-5.3-Flash-Uncensored-FP8 is in the OpenLLM catalog on a dedicated 4× H200 141GB. Deploy it with a flat time pack and call your instance 
- [MiMo-V2.6-Pro-RL-UNCENSORED](https://openllmbuddy.cloud/selfdeploy/self-deploy/models/mimo-v2-6-pro-rl-uncensored): handle `mimo-v2-6-pro-rl-uncensored` — MiMo-V2.6-Pro-RL-UNCENSORED is in the OpenLLM catalog on a dedicated 8× H200. Deploy it with a flat time pack and call your instance through
- [Qwen3.8-27B-Uncensored-FP8](https://openllmbuddy.cloud/selfdeploy/self-deploy/models/qwen3-8-27b-uncensored-fp8): handle `qwen3-8-27b-uncensored-fp8` — Qwen3.8-27B-Uncensored-FP8 is in the OpenLLM catalog on a dedicated H100 80GB. Deploy it with a flat time pack and call your instance throug
- [DeepSeek-V4.1-Flash-UNCENSORED-FP8](https://openllmbuddy.cloud/selfdeploy/self-deploy/models/deepseek-v4-1-flash-uncensored-fp8): handle `deepseek-v4-1-flash-uncensored-fp8` — DeepSeek-V4.1-Flash-UNCENSORED-FP8 is in the OpenLLM catalog on a dedicated 4× H200 141GB. Deploy it with a flat time pack and call your ins
- [Gemma 4 26B A4B](https://openllmbuddy.cloud/selfdeploy/self-deploy/models/gemma-4-26b): handle `gemma4:26b` — Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B ac
- [Qwen3.6 27B A4B](https://openllmbuddy.cloud/selfdeploy/self-deploy/models/qwen-3.6-27b): handle `qwen3.6:27b` — Qwen3.6 27B A4B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026. It features hybrid mul

## Integrations

- [n8n](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/n8n): Automate workflows and call your model as a node.
- [OpenClaw](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/openclaw): Build AI agents and tools on an OpenAI-compatible endpoint.
- [Hermes](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/hermes): Connect agent runners to your chat completions endpoint.
- [OpenCode](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/opencode): Power developer tools with your OpenAI-compatible model.
- [Cursor](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/cursor): Override OpenAI Base URL in Cursor Settings and use your model with BYOK.
- [VS Code](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/vscode): Use the Cline extension in VS Code to connect your OpenAI-compatible endpoint.
- [Codex](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/codex): Run OpenAI Codex CLI against your Chat Completions endpoint via config.toml.
- [Raspberry Pi](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/raspberry-pi): Full Pi OS guide: SSH, API keys, curl, Python venv, systemd, and troubleshooting.
- [Open WebUI](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/open-webui): Self-hosted ChatGPT-style interface with your OpenAI-compatible endpoint as the backend.
- [LibreChat](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/librechat): Multi-provider chat app. Add OpenLLM Buddy as a custom endpoint in librechat.yaml.
- [SillyTavern](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/sillytavern): Roleplay and creative-writing front end. Use the Custom (OpenAI-compatible) chat source.
- [Continue](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/continue): Open-source AI assistant for VS Code and JetBrains. Add a model in config.yaml.
- [Aider](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/aider): Terminal pair programmer that edits your git repo. Point it at your endpoint with two flags.
- [Zed](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/zed): Fast native editor with a built-in AI agent. Add an OpenAI-compatible provider in settings.json.
- [OpenAI SDK (Python & JS)](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/openai-sdk): The official OpenAI client libraries, pointed at your endpoint by changing base_url.
- [LangChain](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/langchain): Use ChatOpenAI with a custom base_url to build chains, tools, and agents on your model.
- [LlamaIndex](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/llamaindex): Use OpenAILike to run RAG pipelines and agents against your OpenAI-compatible model.
- [LiteLLM](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/litellm): One gateway for many providers. Route to your endpoint with the openai/ prefix, SDK or proxy.
- [Dify](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/dify): Build agents, chatbots, and workflows visually. Add OpenLLM Buddy as an OpenAI-compatible provider.
- [Flowise](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/flowise): Drag-and-drop LLM flows. Use the ChatOpenAI Custom node with your base path.
- [Langflow](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/langflow): Visual agent and RAG builder. Point the OpenAI component at your endpoint.
- [Home Assistant](https://openllmbuddy.cloud/selfdeploy/self-deploy/integrations/home-assistant): Run a private voice and chat assistant for your smart home on your own model.

## Optional

- [Blog](https://openllmbuddy.cloud/blog): Articles on deployments, APIs, and GPU inference
- [Support](https://openllmbuddy.cloud/selfdeploy/self-deploy/support): Help with billing, deployments, and API keys
- [Privacy](https://openllmbuddy.cloud/selfdeploy/self-deploy/privacy): Privacy policy
- [Terms](https://openllmbuddy.cloud/selfdeploy/self-deploy/terms): Terms of service
- [Disclaimer](https://openllmbuddy.cloud/selfdeploy/self-deploy/disclaimer): Lawful use and AI output disclaimer