Connect

8 clients · secrets stay placeholders

JSON (mcpServers)Claude CodeCodexOpenCodeCursorVS CodeGemini CLIGoose

{
  "mcpServers": {
    "vram-mcp": {
      "type": "http",
      "url": "https://mcp.nodegrove.io/mcp"
    }
  }
}

claude mcp add --transport http vram-mcp https://mcp.nodegrove.io/mcp

[mcp_servers.vram-mcp]
url = "https://mcp.nodegrove.io/mcp"

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "vram-mcp": {
      "type": "remote",
      "url": "https://mcp.nodegrove.io/mcp",
      "enabled": true
    }
  }
}

{
  "mcpServers": {
    "vram-mcp": {
      "type": "http",
      "url": "https://mcp.nodegrove.io/mcp"
    }
  },
  "deeplink": "cursor://anysphere.cursor-deeplink/mcp/install?name=vram-mcp&config=eyJ0eXBlIjoiaHR0cCIsInVybCI6Imh0dHBzOi8vbWNwLm5vZGVncm92ZS5pby9tY3AifQ%3D%3D"
}

{
  "servers": {
    "vram-mcp": {
      "type": "http",
      "url": "https://mcp.nodegrove.io/mcp"
    }
  }
}

{
  "mcpServers": {
    "vram-mcp": {
      "httpUrl": "https://mcp.nodegrove.io/mcp"
    }
  }
}

extensions:
  "vram-mcp":
    enabled: true
    name: "vram-mcp"
    type: streamable_http
    uri: "https://mcp.nodegrove.io/mcp"
    timeout: 300

Quality 100 / 100strong

protocol100 · weight 25

pass
protocol.modernsupports 2026-07-28 (server/discover)
pass
protocol.statelessanswers without a session
pass
protocol.transportstreamable HTTP
pass
protocol.list_ttltool list is cacheable (ttlMs 3600000)

authorizationn/a · weight 15

n/a
auth.prm, auth.as_metadata, auth.cimdno remote uses OAuth
n/a
auth.secret_in_urlno templated URL

tool hygiene100 · weight 30

pass
tools.descriptionsevery tool has a description
pass
tools.description_lengthdescriptions are concise
pass
tools.schemasevery tool has an object input schema
pass
tools.annotations6 of 6 tools declare readOnlyHint or destructiveHint
pass
tools.directory_hintsevery tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHint
pass
tools.token_costabout 2,723 tokens to load every tool
pass
tools.api_dumptools are not a one-to-one API dump

stability100 · weight 15

pass
stability.changes3 tool changes in 30 days
pass
stability.rug_pullno tool changed its meaning under the same name

dependenciesn/a · weight 15

n/a
deps.known_vulns, deps.mcp_sdk_version, deps.resolvabledependencies not resolved yet

6 tools, about 2,723 tokens to load them all · tool set 16e76380f643 · tools last changed 2026-10-16 · checked 2026-10-20 09:31 UTC

Tool changes

2026-10-16list_modelschangeddescription
before
The open-weight LLMs nodegrove.io has verified against their config.json (data version 2026-10-09): id, size, attention design, native context, licence, memory at Q4 with 8k context and each model's page.
after · 94% of words in common
The open-weight LLMs nodegrove.io has verified against their config.json (data version 2026-10-16): id, size, attention design, native context, licence, memory at Q4 with 8k context and each model's page.
2026-10-09list_modelschangeddescription
before
The open-weight LLMs nodegrove.io has verified against their config.json (data version 2026-10-06): id, size, attention design, native context, licence, memory at Q4 with 8k context and each model's page.
after · 94% of words in common
The open-weight LLMs nodegrove.io has verified against their config.json (data version 2026-10-09): id, size, attention design, native context, licence, memory at Q4 with 8k context and each model's page.
2026-10-05list_modelschangeddescription
before
The open-weight LLMs nodegrove.io has verified against their config.json (data version 2026-09-25): id, size, attention design, native context, licence, memory at Q4 with 8k context and each model's page.
after · 89% of words in common
The open-weight LLMs nodegrove.io has verified against their config.json (data version 2026-10-06): id, size, attention design, native context, licence, memory at Q4 with 8k context and each model's page.

Full history: list_changes · trend: quality JSON.

Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed. How it is computed.

Trust 99 / 100

provenance100

liveness98

freshness100

hygiene100

+repository +namespace-matches-repo +dns-namespace +website +reachable +uptime-95% +updated-0d-ago

Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed. How it is computed.

Dependencies

@nodegrove/[email protected] npmnot resolved yet; dependencies are read once a day

Dependency graphs from deps.dev; advisories from OSV.dev, including the GitHub Advisory Database and the PyPI Advisory Database; all CC BY 4.0. The graph is what a clean install of this version resolves today; nothing was installed or run. How it is checked.

Endpoints

remote	auth	reachable	uptime 30d	p50	protocol	last ok
`https://mcp.nodegrove.io/mcp` streamable-http	none	yes	95%	406 ms	2026-07-28	2026-10-20

package	registry	version	runtime	secrets
`@nodegrove/vram-mcp`	npm	1.1.4	npx

Tools (6)

`can_i_run`read-only	Can this GPU run this open-weight LLM? Returns fits, tight or no, the memory split (weights, KV cache, overhead), a decode-speed ceiling, the longest context that fits and, on a no, every change that would make it fit: quantisation, KV cache, context, another card or a smaller model. Model: a name or id from list_models, any Hugging Face repo id, or its architecture. GPU: a name or id from list_gpus, or vram_gb for any other card.
`estimate_from_hf_repo`read-only	Reads any Hugging Face model repo's config.json and parameter count and estimates its memory: the attention layout found (standard, sliding-window, hybrid or latent), how much each 1,000 tokens of context costs, and weights + KV cache + overhead at every quantisation. For models nodegrove.io has not reviewed; anything the reader cannot model is listed in warnings.
`estimate_vram`read-only	How much memory an LLM needs: weights + KV cache + overhead at each quantisation (or one), at a given context, and the smallest common card class that holds each. Model: a name or id from list_models, any Hugging Face repo id, or its architecture (params_b, layers, kv_heads, head_dim).
`list_gpus`read-only	The GPUs and machines nodegrove.io covers: memory, the memory a runtime can use and bandwidth, from the makers' specs, with each one's page.
`list_models`read-only	The open-weight LLMs nodegrove.io has verified against their config.json (data version 2026-10-16): id, size, attention design, native context, licence, memory at Q4 with 8k context and each model's page.
`what_fits`read-only	Which open-weight LLMs fit this GPU: every model in list_models checked at one quantisation and context, with a recommended everyday model (the biggest class that fits with room for context at conversational speed), the largest that fits, the best at Q8 and the first out of reach. GPU: a name or id from list_gpus, or vram_gb for any other card.

Schemas: list_tools.

Similar MCP servers

Official server.json

{
  "name": "io.nodegrove/vram-mcp",
  "icons": [
    {
      "src": "https://nodegrove.io/icon-512.png",
      "sizes": [
        "512x512"
      ],
      "mimeType": "image/png"
    }
  ],
  "title": "Nodegrove VRAM: can I run it?",
  "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
  "remotes": [
    {
      "url": "https://mcp.nodegrove.io/mcp",
      "type": "streamable-http"
    }
  ],
  "version": "1.1.4",
  "packages": [
    {
      "version": "1.1.4",
      "transport": {
        "type": "stdio"
      },
      "identifier": "@nodegrove/vram-mcp",
      "runtimeHint": "npx",
      "registryType": "npm"
    }
  ],
  "repository": {
    "url": "https://github.com/nodegrove/vram-mcp",
    "source": "github"
  },
  "websiteUrl": "https://nodegrove.io/mcp",
  "description": "Can this LLM run on my GPU? VRAM, speed ceiling and what fits instead, for any model and GPU."
}