Thunder Compute
- 33 repo stars
- Updated July 15, 2026
- Thunder-Compute/thunder-cli
Thunder Compute is a MCP server that lets agents create, manage, and monitor GPU cloud instances, snapshots, SSH keys, and billing on Thunder Compute.
About
Thunder Compute is a Model Context Protocol server that exposes Thunder’s GPU cloud platform to Claude Code, Cursor, Codex, and other MCP-capable agents. Instead of context-switching to a dashboard or CLI alone, developers can ask the agent to spin up instances, check status, manage snapshots, configure SSH access, and review billing in natural language. The server is published as io.github.Thunder-Compute/thunder-compute with a remote streamable-http transport, which suits teams that want a hosted endpoint rather than a local stdio process. It fits ML engineers, inference startups, and agent developers who need occasional high-end GPUs without maintaining their own metal. Pair it with your normal deploy and monitoring stack: this MCP is the control plane hook, not a substitute for app-level observability or security review before you expose training jobs to the internet.
- Streamable HTTP remote MCP at api.thundercompute.com:8443 (server schema 1.0.0)
- Create, manage, and monitor GPU cloud instances from the agent
- Snapshots and SSH key management for reproducible dev environments
- Billing visibility so indie builders can watch GPU spend
- Backed by thunder-cli on GitHub (Thunder-Compute/thunder-cli)
Thunder Compute by the numbers
- Data as of Jul 18, 2026 (Skillselion catalog sync)
claude mcp add --transport http thunder-compute https://api.thundercompute.com:8443/mcpAdd your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| repo stars | ★ 33 |
|---|---|
| Transport | HTTP |
| Auth | None |
| Last updated | July 15, 2026 |
| Repository | Thunder-Compute/thunder-cli ↗ |
What it does
Let your coding agent provision GPU instances, snapshots, SSH keys, and billing on Thunder Compute without leaving the editor.
Who is it for?
Best when you already use Thunder Compute and want agent-driven instance and key management.
Skip if: Skip if you only need local CPU dev boxes or and are not willing to trust an agent with cloud infrastructure actions.
What you get
After you register the remote MCP endpoint, your agent can drive Thunder GPU infrastructure conversations end-to-end with fewer manual CLI and console steps.
- Agent-callable tools for GPU instance create/manage/monitor flows
- SSH key and snapshot operations aligned with Thunder Compute
- Billing and status queries usable inside agent threads
By the numbers
- Server version 1.0.0 in the published manifest
- Remote transport: streamable-http at https://api.thundercompute.com:8443/mcp
- Repository: github.com/Thunder-Compute/thunder-cli
README.md
Overview
tnr is the official command-line interface for Thunder Compute, a high-performance cloud GPU platform built for AI/ML prototyping and experimentation.
Using a proprietary orchestration engine, Thunder Compute delivers fast provisioning, low-latency execution, and one of the most cost-effective GPU compute offerings available.
The tnr CLI supports:
- Provisioning and managing GPU instances
- Configuring compute resources and instance specifications
- Secure SSH access and session management
- File transfer capabilities (SCP upload and download)
- Port forwarding for accessing remote services locally
- Automated update checks and version management
- Cross-platform support for macOS, Linux, and Windows
Documentation
- CLI Reference: https://www.thundercompute.com/docs/cli-reference
- API Reference: https://www.thundercompute.com/docs/api-reference
- Get Started: https://www.thundercompute.com/docs/quickstart
- Troubleshooting: https://www.thundercompute.com/docs/troubleshooting
Installation
Install tnr using one of the supported methods below. You may also download installers and binaries directly from the Latest Release page.
macOS
Download the .pkg installer for your platform from the Latest Release page (tnr_*_darwin_arm64.pkg for Apple Silicon, tnr_*_darwin_amd64.pkg for Intel).
Homebrew:
brew tap Thunder-Compute/tnr
brew install tnr
Linux
Download .tar.gz, .deb, .rpm, or .apk packages from the Latest Release page.
Install script (recommended):
curl -fsSL https://raw.githubusercontent.com/Thunder-Compute/thunder-cli/main/scripts/install.sh | bash
Windows
Download the .msi installer for your platform from the Latest Release page.
Scoop:
scoop bucket add tnr https://github.com/Thunder-Compute/scoop-bucket
scoop install tnr
Winget:
winget install Thunder.tnr
Build from Source
git clone https://github.com/Thunder-Compute/thunder-cli.git
cd thunder-cli
go build -o tnr
./tnr
Quick Start
tnr login # Authenticate with Thunder Compute
tnr create # Create a GPU instance
tnr status # View instance status
tnr connect 0 # Connect to your instance
# File transfers
tnr scp myfile.py 0:/home/ubuntu/
tnr scp 0:/home/ubuntu/results.txt ./
tnr delete 0 # Delete instance
License
This project is licensed under the MIT License. See LICENSE for details.
Recommended MCP Servers
나라장터 낙찰정보나라장터 낙찰정보서비스 Open API를 위한 로컬 MCP 서버나라장터 사전규격나라장터 사전규격정보서비스 Open API를 위한 로컬 MCP 서버나라장터 입찰공고나라장터 입찰공고정보서비스(공공데이터포털 data.go.kr) Open API를 감싼 로컬 MCP 서버
Accelerators36 tools for AWS infrastructure, security compliance, AI workflows, and ACI governance.
ACMSApple Container MCP Server (ACMS) provides access to Apple's container CLI tool on macOS3
Acronis MCPThe first real CLI for the Acronis Cyber Protect Cloud platform - every tenant, agent, and usage15How it compares
Remote GPU infrastructure MCP, not a local coding skill or generic file-editing plugin.
FAQ
Who is Thunder Compute MCP for?
and small-team developers on Thunder’s GPU cloud who use Claude Code, Cursor, or similar agents and want MCP tools for instances, snapshots, SSH keys, and billing.
When should I use Thunder Compute MCP?
Use it during operate-phase infra work—spinning GPUs for training or inference, snapshotting environments, rotating SSH keys, or checking spend—without leaving the agent session.
How do I add Thunder Compute to my agent?
Add the published remote server URL https://api.thundercompute.com:8443/mcp (streamable-http) per your client’s MCP remote config, using the io.github.Thunder-Compute/thunder-compute manifest; confirm auth requirements on Thunder’s docs before production use.