Index / MCP servers / ComputeSage StackBench

ComputeSage StackBench MCP server

ComputeSage StackBench is a remote MCP server that gives an AI agent 5 tools, among them check_deployment_fit, generate_launch_config, predict_performance. It answers at https://mcp.computesage.com/mcp over streamable HTTP, and it listed its tools to an anonymous client, so no sign-in was needed for that. Our probes reached it 10 of 10 times since 4 September 2026; the last check, on 18 October 2026, succeeded.

Status
Answering
Tools
5
Official
yes, served from computesage.com
Reliability
10 of 10 checks
Response time
1105 ms
Transport
streamable HTTP
Last checked
18 October 2026

GPU and LLM inference benchmarks, hardware evidence, deployment recommendations, and launch configs.

Description as the publisher wrote it.

The 5 tools ComputeSage StackBench offers

ToolWhat it does
check_deployment_fit$0.01 USDC on Base mainnet: check model/configuration fit against bounded StackBench capacity evidence.
Inputs: model, hardware_system, quantization, engine, context_length, concurrency
generate_launch_config$0.50 USDC: generate deterministic vLLM or llama.cpp launch templates from an exact StackBench recommendation without claiming a launch occurred.
Inputs: runtime, model, hardware_system, prompt_tokens, quantization, engine_version
predict_performance$0.05 USDC: predict configuration throughput, memory, and power from bounded StackBench evidence with confidence, missing metrics, and caveats.
Inputs: model, hardware_system, quantization, engine, engine_version, kv_cache_type_k
recommend_deployment$0.25 USDC: recommend what to buy or run from bounded StackBench fit, performance, confidence, evidence, caveats, and economics only when supported.
Inputs: model, hardware_system, mode, quantization, engine, engine_version
search_evidence$0.00 free search of bounded public StackBench benchmark and failure evidence; use this for evidence-only requests.
Inputs: query, limit

Add ComputeSage StackBench to your AI client

Every block below points at https://mcp.computesage.com/mcp. Pick your client and copy.

Claude Code: one command in your terminal
claude mcp add --transport http computesage-stackbench https://mcp.computesage.com/mcp
Claude Desktop: claude_desktop_config.json
{
  "mcpServers": {
    "computesage-stackbench": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.computesage.com/mcp"]
    }
  }
}
Cursor: ~/.cursor/mcp.json, or .cursor/mcp.json in a project
{
  "mcpServers": {
    "computesage-stackbench": {
      "url": "https://mcp.computesage.com/mcp"
    }
  }
}
VS Code and GitHub Copilot: .vscode/mcp.json
{
  "servers": {
    "computesage-stackbench": {
      "type": "http",
      "url": "https://mcp.computesage.com/mcp"
    }
  }
}
Zed: Zed settings.json
{
  "context_servers": {
    "computesage-stackbench": {
      "url": "https://mcp.computesage.com/mcp"
    }
  }
}
Windsurf: ~/.codeium/mcp_config.json
{
  "mcpServers": {
    "computesage-stackbench": {
      "serverUrl": "https://mcp.computesage.com/mcp"
    }
  }
}
JetBrains AI Assistant: Settings, Tools, AI Assistant, MCP, New MCP Server, as JSON
{
  "mcpServers": {
    "computesage-stackbench": {
      "url": "https://mcp.computesage.com/mcp"
    }
  }
}
Continue: .continue/mcpServers/neuronto.yaml
name: ComputeSage StackBench
version: 0.0.1
schema: v1
mcpServers:
  - name: ComputeSage StackBench
    type: streamable-http
    url: https://mcp.computesage.com/mcp
Gemini CLI: ~/.gemini/settings.json
{
  "mcpServers": {
    "computesage-stackbench": {
      "httpUrl": "https://mcp.computesage.com/mcp"
    }
  }
}
Codex CLI: MCP server config
{
  "mcpServers": {
    "computesage-stackbench": {
      "type": "http",
      "url": "https://mcp.computesage.com/mcp"
    }
  }
}
Warp: MCP servers, + Add, paste JSON
{
  "mcpServers": {
    "computesage-stackbench": {
      "url": "https://mcp.computesage.com/mcp"
    }
  }
}
Goose: a one-click extension link
goose://extension?url=https%3A%2F%2Fmcp.computesage.com%2Fmcp&type=streamable_http&id=computesage-stackbench&name=ComputeSage%20StackBench

Is ComputeSage StackBench up right now?

It answered our last check on 18 October 2026 in 1105 ms. Over 10 checks since 4 September 2026 it answered 10. Status here comes from Neuronto's own probes, not from the publisher.

MCP servers similar to ComputeSage StackBench

Questions about ComputeSage StackBench

Is there an official ComputeSage StackBench MCP server?

Yes. It is published by computesage.com and served from mcp.computesage.com, the vendor's own domain: https://mcp.computesage.com/mcp.

How many tools does ComputeSage StackBench have?

5, as returned by the server's own tools/list on 10 September 2026: check_deployment_fit, generate_launch_config, predict_performance, recommend_deployment, search_evidence.

Is ComputeSage StackBench down?

It answered our last check on 18 October 2026 in 1105 ms. Over 10 checks since 4 September 2026 it answered 10.

How do I add ComputeSage StackBench to Cursor?

Put the server under mcpServers in ~/.cursor/mcp.json with the URL https://mcp.computesage.com/mcp. The exact JSON is above, with the same setup for Claude Code, Claude Desktop, VS Code, Windsurf, Zed, Gemini CLI and the rest.

Does ComputeSage StackBench need an API key?

It listed its tools to an anonymous client, so reading the tool list needs no key. Individual tools can still ask for credentials in their own inputs.

Where this comes from

Publisher: com.computesage. Endpoint host: mcp.computesage.com. ARD identifier: urn:air:com.computesage:mcp:stackbench. First indexed 31 August 2026. The tools are exactly what the server returned; nothing on this page is inferred from its name. Ranking in Neuronto search is never for sale.

Is this your server? Get a message the day it stops answering or its tools change: monitoring, 14 days free. Or claim the domain to manage how it is listed.