Skip to content

render-examples/anything-llm-render-template

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AnythingLLM on Render

Deploy AnythingLLM, the all-in-one private AI workspace for documents, chat, agents, and RAG, on Render with Mintplex's official :render image and a persistent disk.

Deploy to Render

Render AnythingLLM Docker image Docs

AnythingLLM

What This Template Shows

This repo packages Mintplex's Render-ready Docker image as a one-click Render Blueprint:

Piece Role
AnythingLLM Private AI workspace: RAG, agents, multi-user chat, 130+ providers
mintplexlabs/anythingllm:render Official image built for Render (UI + API + collector)
Render Web Service Runs the image on a Standard instance
Render Disk Persists SQLite, documents, models, and vector cache at /storage

No monorepo build on Render. LLM API keys are configured in the AnythingLLM UI after first login.

Architecture

flowchart LR
  browser["Browser"] --> web["anythingllm<br/>mintplexlabs/anythingllm:render"]
  web --> disk[("/storage disk<br/>SQLite + docs + models")]
  web -.->|"after setup"| llm["Your LLM / embed provider"]
Loading

How It Works

  1. Click Deploy to Render. Render forks this template into your GitHub account and applies render.yaml.
  2. Render pulls mintplexlabs/anythingllm:render-1.15.0, mounts a 10 GB disk at /storage, and starts the container on port 3001.
  3. Open the *.onrender.com URL and create the admin account.
  4. In Settings → LLM Preference, connect OpenAI, Anthropic, Gemini, Ollama, OpenRouter, or another supported provider.
  5. Upload documents and chat. Data stays on the disk across deploys.
Resource Type Plan Notes
anythingllm Web (runtime: image) standard Official Render image; health check /api/ping
anythingllm-storage Disk (10 GB) Mounted at /storage (STORAGE_DIR)

Default region: oregon. This layout uses SQLite on disk, matching Mintplex's upstream render branch Blueprint. It is not a managed Postgres stack.

Quick Start

Prerequisites

  • A Render account
  • An LLM provider key when you are ready to chat (not required at Apply time)

Deploy

  1. Click Deploy to Render above and fork into your GitHub account.
  2. On Apply, confirm the anythingllm service and anythingllm-storage disk. Leave AUTH_TOKEN blank unless you want an instance-wide password gate.
  3. Wait until the service is Live (~3–8 minutes; first image pull can take longer).
  4. Open the public URL and finish first-time setup.
  5. Connect an LLM provider in the UI, upload a document, and start a workspace chat.

Health check:

curl -sS https://<your-service>.onrender.com/api/ping

Features

Feature Description
Official Render image Uses Mintplex's maintained render / render-* tags
One-click Blueprint Web service + disk via projects / environments
Persistent workspace SQLite DB, documents, models, and vector cache on disk
Provider-flexible Configure models in the UI after deploy
Generated secrets JWT_SECRET, SIG_KEY, and SIG_SALT created on first deploy
Pinned by default Image set to render-1.15.0; bump deliberately

Configuration

Variable Source Description
PORT Blueprint 3001 (image listen port)
STORAGE_DIR Blueprint /storage (must match the disk mount)
JWT_SECRET Auto-generated Session / JWT seeding
SIG_KEY Auto-generated Signing key
SIG_SALT Auto-generated Signing salt
AUTH_TOKEN Optional (sync: false) Instance-wide password gate; leave empty for UI auth only

LLM keys (OPEN_AI_KEY, Anthropic, Gemini, etc.) are normally set in the UI. Env injection is documented upstream in docker/.env.example and docs.anythingllm.com/configuration.

Pin or float the image

# render.yaml
image:
  url: docker.io/mintplexlabs/anythingllm:render-1.15.0
  # floating Render channel:
  # url: docker.io/mintplexlabs/anythingllm:render

autoDeployTrigger: off so tag or Blueprint edits do not redeploy until you choose Manual Deploy.

Cost

Resource Approx. monthly
Web service (Standard) ~$25
Disk (10 GB) ~$2.50
Total ~$27–30

LLM tokens are billed by your provider. Do not drop the web plan to Starter (512 MB): the process typically OOMs and Render reports "No open ports detected."

Troubleshooting

Problem Solution
Health check fails / no open ports Use Standard (or higher). Confirm PORT=3001 and disk mounted at /storage.
Reached heap limit Undersized instance. Keep Standard+.
Image pull failures Confirm the tag exists on Docker Hub. Retry the deploy.
Data lost after redeploy Disk must remain attached at /storage. Snapshot before destructive changes.
Login / session weirdness after secret rotate Do not casually rotate JWT_SECRET / SIG_* after users exist.

Project Structure

render.yaml       Render Blueprint (image + disk)
README.md         This file
LICENSE           MIT (template wrapper)
.env.example      Optional env overrides
assets/           Hero / logo placeholders

Learn More

Render:

AnythingLLM:

License

MIT for this template wrapper.

Upstream AnythingLLM is also MIT. Star that repo if this helped.

About

One-click AnythingLLM on Render: official mintplexlabs/anythingllm:render image + disk

Resources

License

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors