MCP Run Python (Pydantic) vs MCP Python Interpreter
Both are data & notebooks MCP servers, and we ran both through the same test (run cell) in each app. MCP Run Python (Pydantic) passed 4 of 4 apps we tested it in; MCP Python Interpreter passed 0 of 3.
Test results side by side
| MCP Run Python (Pydantic) | MCP Python Interpreter | |
|---|---|---|
| Claude Code | Passed 2026-10-15 | Failed 2026-10-15 |
| Cursor | Passed 2026-10-15 | Failed 2026-10-15 |
| Claude Desktop | Passed 2026-10-18 | Not tested yet |
| ChatGPT (Codex) | Passed 2026-10-20 | Failed 2026-10-22 |
| Tools exposed (protocol check) | 1 tools | — |
| GitHub stars | 194 | 104 |
| License | MIT | NOASSERTION |
| Last push | 2026-01-30 | 2026-07-02 |
Results from our own harness runs, dated per cell. Stars and last push from GitHub at our last data refresh. We don’t pick a winner here: which one fits depends on your app and setup.
What each does
MCP Run Python (Pydantic): Runs Python code in a sandbox (Pyodide on Deno) and returns the output. Needs Deno installed. The maker archived the repo.
MCP Python Interpreter: Runs Python code and manages Python environments and packages from an agent, inside a working folder.
