A side-by-side look at the Hub MCP and Nodegrove VRAM: can I run it? MCP servers: transport, auth, tools, quality signals, and the config to install either one in Claude, Cursor, Windsurf or VS Code.
Verdict: choose Hub MCP for developer tools work over a hosted endpoint; choose Nodegrove VRAM: can I run it? for developer tools as a local process. You can also run both side by side in one mcpServers block.
HUB_PAT_TOKEN, MCP_AUTH_TOKEN.search, query, architectures.can_i_run, what_fits, estimate_vram.Rows where the two servers differ are marked.
| Spec | N Nodegrove VRAM: can I run it? | |
|---|---|---|
| Category | Developer Tools | Developer Tools |
| Quality signal | 77/100 · Great | 53/100 · Good |
| Install command | Confirmed | Confirmed |
| Transport | Remote HTTP/SSE | Local subprocess (stdio) |
| Auth | API key required | No auth required |
| Pricing | Free | Free |
| Required env vars | HUB_PAT_TOKENMCP_AUTH_TOKEN | None listed |
| Verified owner | Verified | Community listing |
| GitHub stars | 170 | 0 |
| Directory activity | 38 views 0 installs 0 upvotes | 0 views 0 installs 0 upvotes |
| Tools | 35 | 4 |
| Full listing | View Hub MCP | View Nodegrove VRAM: can I run it? |
searchSearch repositories and content using Search V4 APIquerySearch query parameter (string, required)architecturesFilter on architectures (string, optional)badgesFilter by image content type badges (string, optional)categoriesFilter on categories (string, optional)extension_reviewedFilter on reviewed extensions (boolean, optional)fromNumber of documents to skip for pagination (number, optional)imagesFilter on image names (string, optional)operating_systemsFilter on operating systems (string, optional)orderChange the ordering of results (string, optional)sizeMaximum number of results to return (number, optional)sortSort results by search field (string, optional)can_i_runFits, tight or no; the memory split; a speed ceiling; the longest context that fits; and on a no, every change that would make it fitwhat_fitsEvery model checked on one card, with a recommended everyday pick, the largest that fits, the best at Q8 and the first out of reachestimate_vramWeights, KV cache and overhead at each quantisation, and the smallest common card that holds eachestimate_from_hf_repoAny Hugging Face repo, read from its `config.json`: the attention layout, the cost of each 1,000 tokens of context, and memory at every quantisationPaste either block (or both) into your client config, e.g. claude_desktop_config.json or ~/.cursor/mcp.json. We only show a config when the install command is confirmed.
{
"mcpServers": {
"docker-hub-mcp": {
"url": "https://docs.docker.com/ai/gordon/mcp/yaml/"
}
}
}
{
"mcpServers": {
"nodegrove-vram-can-i-run-it": {
"command": "npx",
"args": [
"-y",
"@nodegrove/vram-mcp"
]
}
}
}