Matrix Compute MCP

● Active

dev.pages.matrix-compute-website/matrix-compute-mcp

P2P GPU matrix multiply, exact cosine search, and PCA/SVD verified via Freivalds' algorithm.

Connect this server

One click to install, or copy the config below. Suits: Claude Desktop · Cursor.

Add to Cursor Add to VS CodeClaude Desktop / Claude Code: paste the config below.

Or add it manually — claude_desktop_config.json (Claude Desktop) or .cursor/mcp.json (Cursor):

{
  "mcpServers": {
    "matrix-compute-mcp": {
      "command": "uvx",
      "args": [
        "matrix-compute-mcp"
      ],
      "env": {
        "MATRIX_NWC_URI": "<your MATRIX_NWC_URI>"
      }
    }
  }
}
One line, from the terminal
Claude Code
claude mcp add --env MATRIX_NWC_URI='<your MATRIX_NWC_URI>' --transport stdio matrix-compute-mcp -- uvx matrix-compute-mcp
OpenAI Codex CLI
codex mcp add matrix-compute-mcp --env MATRIX_NWC_URI='<your MATRIX_NWC_URI>' -- uvx matrix-compute-mcp
VS Code
code --add-mcp '{"name":"matrix-compute-mcp","type":"stdio","command":"uvx","args":["matrix-compute-mcp"],"env":{"MATRIX_NWC_URI":"<your MATRIX_NWC_URI>"}}'

Syntax from each client's docs (Claude Code: claude mcp add; Codex: codex mcp add; VS Code: code --add-mcp). Placeholders in angle brackets are values you fill in.

Environment variables
  • MATRIX_NWC_URIsecretOptional Nostr Wallet Connect (nostr+walletconnect://) URI for mainnet jobs

A starting point — check the repository for exact arguments and setup. Restart your client after editing the config.

Category
Communication
Version
0.1.3
Registry
active
Published
11/15/2026
Listed in:Official MCP registry ↗how this is checked

Is it safe? Is it maintained?

Serverdex check: Nov 22, 2026

Looks healthy: Matrix Compute MCP is maintained, open source, hosted endpoint answering without a key.

Maintained
Active
Last commit 7 days ago (Nov 15, 2026), 0 open issues.
Official registry
Active
Published as dev.pages.matrix-compute-website/matrix-compute-mcp in the official MCP registry, version 0.1.3.
Source
Open source, Apache-2.0
Public repository under the Apache-2.0 licence, 0 stars, 0 forks.
Runs
Local package or hosted
Ships as a local package (PyPI) that runs on your machine, and as a hosted endpoint at mcp.matrix-compute.net that receives what your client sends it.
Credentials
None needed to connect
The hosted endpoint completed the Serverdex initialize + tools/list handshake with no key (checked Nov 22, 2026); individual tools may still ask for one.
Owner
Unclaimed
Nobody has claimed this page yet; the data above comes from the registry, GitHub and the live check.

Facts only: last commit and licence from GitHub, the entry's status from the official registry, credentials from the live check. Not a code audit — run the security scan on the endpoint yourself and see the status history.

Live check

checked Nov 22, 2026, 12:07 UTCStatus history & uptime →
● Working209 msprotocol 2024-11-05v0.1.0toolsresourcesprompts

The endpoint answered the MCP handshake as "matrix-compute-mcp" and listed 5 tools. Answering since Nov 18, 2026.

Context cost ≈ 1.2k tokens of tool definitions a client loads per conversation (estimate: characters ÷ 4) across 5 tools.

Tools (5)

  • get_network_status2 params

    Queries the Matrix Compute discovery relay for live GPU worker availability, total VRAM capacity, confidential computing (TEE) enclaves, and sats_per_gflop pricing.

    Show parameters
    • min_vram_mbintegerMinimum worker VRAM filter in MB (optional).
    • max_price_per_gflopnumberMaximum worker price filter in satoshis per GFLOP (optional).
  • estimate_job_cost10 params

    Calculates exact upload/download network bandwidth (MB), task count, GFLOPs, and satoshi cost estimates for a distributed matrix multiplication C (m x n) = A (m x k) x B (k x n).

    Show parameters
    • mintegerrequiredRows of matrix A (M > 0).
    • kintegerrequiredInner dimension K (columns of A / rows of B).
    • nintegerrequiredColumns of matrix B (N > 0).
    • dtypestringPrecision format ('fp32', 'fp16', 'bf16').
    • partitioning_strategystringMatrix partitioning strategy.
    • tile_sizeintegerTile block size dimension.
    • num_stripsintegerStrip count when using outer_product_1d.
    • grid_rowsintegerRow partitions P_r when using output_grid_2d.
    • grid_colsintegerColumn partitions P_c when using output_grid_2d.
    • sats_per_gflopnumberOverride worker price in satoshis per GFLOP.
  • execute_distributed_matmul8 params

    Executes verified distributed matrix multiplication C = A x B across decentralized GPU workers with dynamic Freivalds verification and L402 settlement. Accepts file paths (.npy, .npz, .safetensors) or inline 2D JSON arrays.

    Show parameters
    • matrix_arequiredFile path (.npy/.npz/.safetensors) or inline 2D array for matrix A (M x K).
    • matrix_brequiredFile path (.npy/.npz/.safetensors) or inline 2D array for matrix B (K x N).
    • output_pathstringOptional output file path (.npy, .npz, or .safetensors) to write C.
    • tensor_key_astringOptional tensor key when loading matrix_a from .safetensors or .npz.
    • tensor_key_bstringOptional tensor key when loading matrix_b from .safetensors or .npz.
    • dtypestringComputation precision ('fp32', 'fp16', 'bf16').
    • partitioning_strategystringMatrix partitioning topology.
    • tile_sizeintegerOptional tile block size override.
  • batch_similarity_search5 params

    Runs exact cosine similarity search between query embeddings (M x D) and corpus embeddings (N x D) on distributed GPUs using OUTPUT_GRID_2D zero-reduction tiling.

    Show parameters
    • queriesrequiredFile path (.npy/.npz/.safetensors) or inline 2D array of shape (M, D).
    • corpus_embeddingsrequiredFile path (.npy/.npz/.safetensors) or inline 2D array of shape (N, D).
    • top_kintegerNumber of top matching corpus indices and scores to return per query.
    • dtypestringTarget GEMM precision.
    • output_pathstringOptional file path (.npy/.safetensors) to save the full similarity matrix.
  • compute_pca_svd5 params

    Computes verified top-k Principal Component Analysis (PCA) / Singular Value Decomposition (SVD) on distributed GPU workers via randomized subspace iteration or stateful power iteration.

    Show parameters
    • matrixrequiredFile path (.npy/.npz/.safetensors) or inline 2D array of shape (M, N).
    • kintegerrequiredNumber of top singular/principal components to extract.
    • centerbooleanMean-center columns before decomposition (standard PCA).
    • methodstringDecomposition algorithm ('randomized' or 'power_iteration').
    • output_pathstringOptional output file path (.npz or .safetensors) to save U, S, Vt.

Checked by Serverdex with a plain initialize + tools/list call to mcp.matrix-compute.net/mcp, no credentials, about once a day. A failed check is never a removal.

On GitHub

read Nov 17, 2026
Stars
0
Forks
0
Last push
Nov 15, 2026
License
Apache-2.0

Python

From the README

matrix-compute-mcp: Official MCP Server & High-Level Verifiable AI Primitives matrix-compute-mcp is an official Model Context Protocol (MCP) server and high-level AI primitives package built on top of matrix-compute-sdk that provides: 1. High-Level Verifiable AI Primitives (matrixcomputemcp.ai): Drop-in PyTorch layers (DistributedLinear, DistributedLoRALinear), massive exact cosine similarity search (distributedcosinesimilarity), and distributed spectral PCA/SVD (distributedpcasvd) — all verified in $O(n^2)$ via Freivalds' algorithm. 2. Official Model Context Protocol Server (matrix-compute-mcp): Exposes decentralized GPU TFLOPS to autonomous AI agents (Claude Desktop, Cursor, Windsurf, Antigravity, LangChain, OpenClaw) over stdio (uvx matrix-compute-mcp) or remote Streamable HTTP (https://mcp.matrix-compute.net/mcp) with L402 Macaroon, NWC (nostr+walletconnect://), or zero-drain trial sandbox support. Quick Start 1A. Local stdio MCP Client Configuration (Claude Desktop / Cursor / Antigravity) 1B. Hosted Remote MCP Endpoint (Zero Install) Point any Streamable HTTP MCP client directly to: 2. Exposed MCP Tools 3. High-Level Verifiable AI Primitives (matrixcomputemcp.ai)

Read the full README on GitHub →

Remotes

  • streamable-httphttps://mcp.matrix-compute.net/mcp

The most-starred Communication servers in the index, with their live-check result.

ServerStarsLast pushLive check
Inbox Zero13,088Nov 2026Needs sign-in
Agent-Native Brain7,735Nov 2026Needs sign-in
Agent-Native Mail7,735Nov 2026Needs sign-in
Agent-Native Chat7,735Nov 2026Needs sign-in
Agent-Native Dispatch7,735Nov 2026Needs sign-in

Matrix Compute MCP: questions people ask

Is Matrix Compute MCP a remote or a local MCP server?
According to its entry in the official MCP registry, Matrix Compute MCP ships as a hosted endpoint (streamable-http) that your client connects to by URL and as a local package (PyPI) that your client starts on your machine. Use the hosted endpoint when your client supports Streamable HTTP; use the package when you want it to run locally.
Does Matrix Compute MCP need an API key?
Its hosted endpoint completed the Serverdex live check — an MCP initialize + tools/list call — with no credentials on Nov 22, 2026. Individual tools may still ask for a key when called.
Is Matrix Compute MCP safe to use?
Looks healthy: Matrix Compute MCP is maintained, open source, hosted endpoint answering without a key. In detail — maintained: Active; official registry: Active; source: Open source, Apache-2.0; runs: Local package or hosted; credentials: None needed to connect. Serverdex check of Nov 22, 2026. This is a reading of the repository, the registry entry and the live check, not a code audit; run the Serverdex security scan on the endpoint before trusting it with real data.
Is Matrix Compute MCP still maintained?
Active: the repository's last commit 7 days ago. Last push Nov 15, 2026, 0 GitHub stars, Apache-2.0 licence. Read from GitHub on Nov 17, 2026.
What tools does Matrix Compute MCP expose?
5 tools as listed by the server itself during the live check, including get_network_status, estimate_job_cost, execute_distributed_matmul, batch_similarity_search, compute_pca_svd. The full list with parameters is in the Live check section above.
How do I add Matrix Compute MCP to Claude, Cursor or VS Code?
Copy the config in "Connect this server" into your client's MCP config file (claude_desktop_config.json, ~/.cursor/mcp.json or .vscode/mcp.json), or use the one-click Add to Cursor / Add to VS Code buttons and the one-line commands for Claude Code and Codex; then restart the client.

More from matrixmultiplyconsortium

All 2 servers by matrixmultiplyconsortium →
  1. Matrix Compute MCPP2P GPU matrix multiply, exact cosine search, and PCA/SVD verified via Freivalds' algorithm.
Browse all Communication →

Add the badge to your site

Show visitors that Matrix Compute MCP is listed here — it’s also the link we check for a free listing.

Matrix Compute MCP — listed on Serverdex

HTML — for your website

<a href="https://serverdex.io/server/dev-pages-matrix-compute-website-matrix-compute-mcp?utm_source=badge"><img src="https://serverdex.io/brand/badge.svg" alt="Matrix Compute MCP — listed on Serverdex" width="123" height="40" loading="lazy"></a>

Recommended. A dofollow link from your own domain is the one that counts for search engines, and it's the link we check for a free listing.

Markdown — for your README

[![Matrix Compute MCP — listed on Serverdex](https://serverdex.io/brand/badge.svg)](https://serverdex.io/server/dev-pages-matrix-compute-website-matrix-compute-mcp?utm_source=badge)

Nice for visibility, but GitHub marks README links nofollow, so it doesn't pass authority. Use the HTML badge on your site as well.

Add to CursorAdd to VS CodeOpen in Claude Code

Install buttons — for your README

[![Add to Cursor](https://serverdex.io/brand/install-cursor.svg)](https://serverdex.io/install/cursor/dev-pages-matrix-compute-website-matrix-compute-mcp) [![Add to VS Code](https://serverdex.io/brand/install-vscode.svg)](https://serverdex.io/install/vscode/dev-pages-matrix-compute-website-matrix-compute-mcp) [![Open in Claude Code](https://serverdex.io/brand/install-claude-code.svg)](https://serverdex.io/install/claude-code/dev-pages-matrix-compute-website-matrix-compute-mcp)

One click opens Cursor or VS Code with this server's config prefilled; the Claude Code button lands on the one-line command. The links point at Serverdex, so when your config changes the buttons keep working.

Install buttons — HTML for your docs

<a href="https://serverdex.io/install/cursor/dev-pages-matrix-compute-website-matrix-compute-mcp"><img src="https://serverdex.io/brand/install-cursor.svg" alt="Add to Cursor — via Serverdex" width="140" height="28"></a>
<a href="https://serverdex.io/install/vscode/dev-pages-matrix-compute-website-matrix-compute-mcp"><img src="https://serverdex.io/brand/install-vscode.svg" alt="Add to VS Code — via Serverdex" width="147" height="28"></a>
<a href="https://serverdex.io/install/claude-code/dev-pages-matrix-compute-website-matrix-compute-mcp"><img src="https://serverdex.io/brand/install-claude-code.svg" alt="Open in Claude Code — via Serverdex" width="180" height="28"></a>

Same buttons as plain links for a docs page or website.

Prefer a live badge with the score, stars and live check? Score badge (links to a page that explains the number and back to this listing).