Evidence record / pi-coding-agent

A dated record of one change, kept so the writing that cites it can be checked. Compiled from the sources listed below by the research run, not written for reading. The judgment lives in the issues that cite it, below.

NVIDIA NIM provider dynamic model discovery

What Changed

Integrated NVIDIA NIM (NVIDIA Inference Microservice) as OpenAI-compatible provider. Fetches available models from https://integrate.api.nvidia.com/v1 API with filtering for text input/output and tool use support. Uses NVIDIA_API_KEY environment variable. Blocks ~14 models incompatible with tool use.

Operator Implication

Operators can now use NVIDIA’s public inference models through unified pi interface. Dynamic model discovery reduces maintenance burden as NVIDIA adds models.

Receipt

Commit 6014801 ‘Add NVIDIA NIM provider’ adds dynamic model discovery from NVIDIA NIM API with tool-use filtering and NVIDIA_API_KEY support

Finding metadata

Confidence: high
Actionability: test
Accessibility impact: low

Run: 2026-06-03-weekly-digest-2026-05-28_2026-06-03-frontier-v0

Finding ID: 2026-06-02-pi-coding-agent-provider-nvidia-nim

Source links

Primary links, including exact changelog lines when available.

Pi Coding Agent
commitCommit 6014801 'Add NVIDIA NIM provider' adds dynamic model discovery from NVIDIA NIM API with tool-use filtering and NVIDIA_API_KEY supportgithub.com/earendil-works/pi/commit/6014801

Versioned source: run artifact