gpu.nz GPU Cloud
Find GPU servers with live per-minute prices, estimate how much VRAM an LLM or image model needs, and launch, monitor and stop GPU servers (ComfyUI, Ollama, vLLM, Jupyter) on the gpu.nz cloud. Browsing and VRAM estimates need no key; renting needs an API key. Endpoint: https://gpu.nz/mcp · Docs
Live Answered an MCP handshake without credentials and listed 16 tools.
- Category
- Cloud, ops & data
- Transport
http- Source
- https://github.com/gpunz/gpunz-mcp
- Endpoint
https://gpu.nz/mcp- MCP Registry
nz.gpu/gpu-cloud- Reports itself as
- gpu.nz 1.0.0
- Protocol
- 2025-06-18
- Tools
- 16
- Context cost
- about 2,360 tokens for the tool definitions (estimated at four characters per token)
- Tool definitions last changed
- no change seen since first check
- Handshake time
- 629 ms
- Repository
- 1 stars · last push 2026-10-21 · MIT
- Last checked
- 2026-11-03 10:54:42 UTC
- Listed since
- 2026-10-20
Health and compatibility
| Endpoint reachable | Verified | Responded in 629 ms |
|---|---|---|
| MCP handshake (initialize) | Verified | Completed without credentials |
| Tool discovery (tools/list) | Verified | 16 tools listed |
| Authentication | Verified | None needed to connect |
| Protocol version | Verified | 2025-06-18 |
| Transport | Verified | Streamable HTTP |
| Capabilities declared | Verified | tools |
Last checked . Verified means the check saw it happen; it says nothing about whether the server is safe.
Recent history
100% of the last 10 daily checks reachable; median response 632 ms. Green answered, violet asked for auth, red did not answer.
Set up in your client
Claude Code
Run this in a terminal:
claude mcp add --transport http gpu-nz https://gpu.nz/mcpClaude Desktop
Open Settings → Connectors → Add custom connector and paste this URL:
https://gpu.nz/mcpOr bridge it through mcp-remote in claude_desktop_config.json (needs Node.js):
{
"mcpServers": {
"gpu-nz": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://gpu.nz/mcp"
]
}
}
}Cursor
Or add it to ~/.cursor/mcp.json (or .cursor/mcp.json in a project):
{
"mcpServers": {
"gpu-nz": {
"url": "https://gpu.nz/mcp"
}
}
}VS Code
Or add it to .vscode/mcp.json:
{
"servers": {
"gpu-nz": {
"type": "http",
"url": "https://gpu.nz/mcp"
}
}
}Or from a terminal:
code --add-mcp '{"name": "gpu-nz", "type": "http", "url": "https://gpu.nz/mcp"}'Windsurf
Add it to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"gpu-nz": {
"serverUrl": "https://gpu.nz/mcp"
}
}
}Other clients
Most clients accept this shape. The endpoint is:
https://gpu.nz/mcp{
"mcpServers": {
"gpu-nz": {
"type": "http",
"url": "https://gpu.nz/mcp"
}
}
}If it does not connect
- Restart the client after editing a config file, and keep keys out of files you commit.
- If the tools do not appear, open the client's MCP log: a 401 means the credentials were not accepted, and a timeout means the server did not answer.
Tools 16
Names, descriptions and input schemas exactly as the server returned them on 2026-11-03. This is the server author's own text. Nothing is run from this page. Labels such as read-only are declared by the server and are not verified. Raw JSON
add_ssh_keycan modifyexternal accessAdds an SSH public key (a line from a .pub file) to the account. Needs user:write.
Adds an SSH public key (a line from a .pub file) to the account. Needs user:write.
| Parameter | Type | Description | |
|---|---|---|---|
public_key | string | required | The public key line, e.g. ssh-ed25519 AAAA… you@laptop |
name | string | optional | A name for the key |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "add_ssh_key",
"arguments": {
"public_key": "..."
}
}
}create_instancecan modifyexternal accessRents the server of an offer and starts it with a template. Billing starts right away and runs every minute …
Rents the server of an offer and starts it with a template. Billing starts right away and runs every minute until you stop or destroy it. Confirm the price with the user first. Needs instances:write. The response has the instance id; poll get_instance until status is running (usually 1–5 minutes).
| Parameter | Type | Description | |
|---|---|---|---|
offer_id | string | required | Offer id from search_offers |
template | string | required | Template slug from list_templates, e.g. "comfyui", "ollama", "pytorch", "custom" |
disk_gb | integer | required | Disk size, GB (use the template's recommended_disk_gb) |
name | string | optional | A name for the server |
type | string | optional | on_demand (default) or interruptible One of: on_demand, interruptible |
image | string | optional | Docker image, only with the custom template |
ssh_key_ids | string[] | optional | SSH key ids to install (default: all of yours) |
idempotency_key | string | optional | Reuse the same value when retrying the same request, so a retry never rents twice |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "create_instance",
"arguments": {
"offer_id": "...",
"template": "...",
"disk_gb": 0
}
}
}destroy_instancecan modifydestructiveexternal accessDestroys a server and deletes its disk for good; billing stops completely. Irreversible: confirm with the …
Destroys a server and deletes its disk for good; billing stops completely. Irreversible: confirm with the user. Needs instances:write.
| Parameter | Type | Description | |
|---|---|---|---|
instance_id | string | required | Instance id |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "destroy_instance",
"arguments": {
"instance_id": "..."
}
}
}estimate_vramread-onlyHow much GPU memory an AI model needs (LLMs such as Qwen, DeepSeek, Llama, Gemma, GPT-OSS; image and video …
How much GPU memory an AI model needs (LLMs such as Qwen, DeepSeek, Llama, Gemma, GPT-OSS; image and video models such as FLUX, SDXL, Wan) at a given precision and context length, and which GPUs on gpu.nz fit it with their current price. Call without `model` to list the known models.
| Parameter | Type | Description | |
|---|---|---|---|
model | string | optional | Model name or id, e.g. "qwen3-32b", "Llama 3.3 70B", "flux.1-dev" |
precision | string | optional | Weights precision: "fp16", "fp8", "q8", "q6", "q5", "q4" (LLMs; default q4 for LLMs, fp16 for image models) |
context_tokens | integer | optional | LLM context length in tokens (default 8192) |
concurrent_requests | integer | optional | Parallel sequences sharing the GPU (default 1) |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "estimate_vram",
"arguments": {}
}
}get_accountread-onlyexternal accessThe signed-in account and its balance, hourly burn and runway. Needs a key with user:read and billing:read.
The signed-in account and its balance, hourly burn and runway. Needs a key with user:read and billing:read.
Takes no parameters.
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "get_account",
"arguments": {}
}
}get_instanceread-onlyexternal accessStatus, SSH command, app URLs (ComfyUI, Jupyter, chat…), price and cost so far of one server. The root …
Status, SSH command, app URLs (ComfyUI, Jupyter, chat…), price and cost so far of one server. The root password is left out unless include_password is true (your SSH keys are installed on every server).
| Parameter | Type | Description | |
|---|---|---|---|
instance_id | string | required | Instance id |
include_password | boolean | optional | Include the root SSH password |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "get_instance",
"arguments": {
"instance_id": "..."
}
}
}get_instance_logsread-onlyexternal accessThe last lines of a server's container log: boot progress, app start-up, errors. Needs instances:read.
The last lines of a server's container log: boot progress, app start-up, errors. Needs instances:read.
| Parameter | Type | Description | |
|---|---|---|---|
instance_id | string | required | Instance id |
tail | integer | optional | How many last lines (default 200) |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "get_instance_logs",
"arguments": {
"instance_id": "..."
}
}
}get_offerread-onlyexternal accessDetails and the current price of one offer by id (from search_offers).
Details and the current price of one offer by id (from search_offers).
| Parameter | Type | Description | |
|---|---|---|---|
offer_id | string | required | Offer id |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "get_offer",
"arguments": {
"offer_id": "..."
}
}
}list_gpu_modelsread-onlyexternal accessGPU models available on gpu.nz right now with VRAM, class and the cheapest hourly price per GPU. Use it to …
GPU models available on gpu.nz right now with VRAM, class and the cheapest hourly price per GPU. Use it to pick a gpu_slug for search_offers.
| Parameter | Type | Description | |
|---|---|---|---|
min_vram_gb | number | optional | Only models with at least this much VRAM per GPU |
query | string | optional | Filter by name, e.g. "4090" or "H100" |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "list_gpu_models",
"arguments": {}
}
}list_instancesread-onlyexternal accessYour rented servers with status, GPU, price and SSH command. Needs instances:read.
Your rented servers with status, GPU, price and SSH command. Needs instances:read.
| Parameter | Type | Description | |
|---|---|---|---|
status | string | optional | active (default): servers that are not destroyed; all: include destroyed ones One of: active, all |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "list_instances",
"arguments": {}
}
}list_ssh_keysread-onlyexternal accessSSH public keys on the account; they are installed on new servers. Needs user:read.
SSH public keys on the account; they are installed on new servers. Needs user:read.
Takes no parameters.
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "list_ssh_keys",
"arguments": {}
}
}list_templatesread-onlyexternal accessReady environments a server can start with: ComfyUI, Stable Diffusion Forge, Ollama + Open WebUI, vLLM, …
Ready environments a server can start with: ComfyUI, Stable Diffusion Forge, Ollama + Open WebUI, vLLM, Jupyter/PyTorch, Kohya, FluxGym, Unsloth, chats with preloaded models, plain Ubuntu + CUDA, or a custom Docker image. Pass `slug` as `template` to create_instance.
Takes no parameters.
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "list_templates",
"arguments": {}
}
}search_offersread-onlyexternal accessSearch rentable GPU servers (offers) with live prices. Returns the cheapest matches first unless `order` says …
Search rentable GPU servers (offers) with live prices. Returns the cheapest matches first unless `order` says otherwise. Each result has an offer id for create_instance.
| Parameter | Type | Description | |
|---|---|---|---|
gpu | string | optional | GPU model slug(s), comma-separated, e.g. "rtx-4090-24gb" or "h100-sxm-80gb,h100-pcie-80gb" (see list_gpu_models) |
num_gpus | string | optional | GPUs per server, comma-separated, e.g. "1" or "2,4,8" |
min_vram_gb | number | optional | Minimum VRAM per GPU, GB |
max_price_rub_h | number | optional | Maximum price for the whole server, rubles per hour |
country | string | optional | ISO-2 country codes, comma-separated, e.g. "DE,NL,FI" |
min_reliability | number | optional | Minimum host reliability, 0–1 (0.98 is a good bar) |
verified | boolean | optional | Only verified datacenters |
min_disk_gb | number | optional | Minimum disk the server can give, GB |
type | string | optional | on_demand (default) or interruptible (cheaper, can be preempted) One of: on_demand, interruptible |
order | string | optional | Sort: price (default), dlperf_per_price for value, reliability, vram, inet_down One of: price, -price, dlperf, dlperf_per_price, reliability, vram, inet_down |
limit | integer | optional | How many offers to return (default 10) |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "search_offers",
"arguments": {}
}
}start_instancecan modifyexternal accessStarts a stopped server again; GPU billing resumes. Confirm with the user. Needs instances:write.
Starts a stopped server again; GPU billing resumes. Confirm with the user. Needs instances:write.
| Parameter | Type | Description | |
|---|---|---|---|
instance_id | string | required | Instance id |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "start_instance",
"arguments": {
"instance_id": "..."
}
}
}stop_instancecan modifyexternal accessStops a running server: the GPU is released and billing drops to the disk only. Data on the disk is kept and …
Stops a running server: the GPU is released and billing drops to the disk only. Data on the disk is kept and the server can be started again (the GPU may then be taken by someone else; gpu.nz finds a replacement when it is). Needs instances:write.
| Parameter | Type | Description | |
|---|---|---|---|
instance_id | string | required | Instance id |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "stop_instance",
"arguments": {
"instance_id": "..."
}
}
}topup_linkread-onlyA link to the balance top-up page (SBP, crypto) for the user to open in a browser; payments never happen …
A link to the balance top-up page (SBP, crypto) for the user to open in a browser; payments never happen through the API. Optionally pre-fills the amount.
| Parameter | Type | Description | |
|---|---|---|---|
amount | integer | optional | Amount in the account currency (rubles by default) |
language | string | optional | Page language (default ru) One of: ru, en |
Example call (required arguments only, placeholder values):
{
"method": "tools/call",
"params": {
"name": "topup_link",
"arguments": {}
}
}Badge
Maintain this server? Put its live status in your README:
[](https://mcp.aianytime.net/s/gpu-nz/)Edit this entry · Report a problem
More in Cloud, ops & data
Checks record what a server answered, not whether it is safe. A listing is not a security audit; read a server's source and permissions before you connect it. Add a server · JSON · llms.txt · llms-full.txt
Browse: Agents, IDEs & developer tools · Search & discovery · Knowledge, memory & context · Media & generation · Security · Finance & markets · Marketing, content & social · Cloud, ops & data · Commerce, travel & logistics · Work & productivity · No account needed · Local servers