rmerk bolts NVIDIA onto Mike and quietly changes the default brain

A fourth AI provider lands, and fresh installs now think with NVIDIA-hosted models instead of Google's Gemini.

infrastructureintegration

rmerk added NVIDIA's hosted AI service as a fourth engine option, alongside the providers Mike already supports. It speaks the same common API dialect everyone else uses, so it slots in cleanly without stepping on the existing options.

The bigger story is the default. A brand-new deployment with no keys configured now reaches for NVIDIA's catalogue rather than Google's Gemini - worth knowing if you're inheriting an environment set up for the old default, because the AI simply won't answer until the right key is in place. Two practical notes: entering a key per-user through the interface isn't wired up yet (it's environment-level only for now, and the team flags this openly), and the same setup can be pointed at a self-hosted or local model instead of NVIDIA's cloud - handy if you'd rather keep everything in-house.

So what Anyone running or evaluating this fork should check which AI provider their deployment actually defaults to before going live.

View this fork on GitHub →

Spotted something wrong? Or know the PR text has fresher detail than the writeup above?

Commits in this thread

1 commit from rmerk/mike, oldest first. Source extracted verbatim from the harvested git log.

SHA Subject Author Date

8535292c feat: add NVIDIA API Catalog provider with Kimi K2.6 as default Ryan Choi 2026-05-11 ↗ GitHub

SHA	Subject	Author	Date
`8535292c`	feat: add NVIDIA API Catalog provider with Kimi K2.6 as default	Ryan Choi	2026-05-11	↗ GitHub
commit body Adds a fourth LLM provider that targets build.nvidia.com's OpenAI-compatible Chat Completions endpoint, enabling Kimi K2.6, Llama 3.3 70B, and DeepSeek R1 through a single env key. Backend: new lib/llm/nvidia.ts handles streaming + multi-turn tool calls against integrate.api.nvidia.com/v1. ApiKeyProvider extended with "nvidia" (env-only - DB CHECK constraint still blocks per-user storage). Frontend: model picker now shows an NVIDIA group at the top with Kimi K2.6 selected by default. ApiKeyState and modelAvailability extended to recognise nvidia so the picker doesn't grey out available models. All gemini-3-flash-preview fallbacks swapped to moonshotai/kimi-k2.6 so the app boots usefully without a Gemini key.

commit body

Adds a fourth LLM provider that targets build.nvidia.com's OpenAI-compatible
Chat Completions endpoint, enabling Kimi K2.6, Llama 3.3 70B, and DeepSeek R1
through a single env key.

Backend: new lib/llm/nvidia.ts handles streaming + multi-turn tool calls
against integrate.api.nvidia.com/v1. ApiKeyProvider extended with "nvidia"
(env-only - DB CHECK constraint still blocks per-user storage).

Frontend: model picker now shows an NVIDIA group at the top with Kimi K2.6
selected by default. ApiKeyState and modelAvailability extended to recognise
nvidia so the picker doesn't grey out available models.

All gemini-3-flash-preview fallbacks swapped to moonshotai/kimi-k2.6 so the
app boots usefully without a Gemini key.

Capture this thread into my fork

Download a single Markdown prompt that tells Claude how to port every commit above into your working tree — adapting paths and structure to match your repo. Run it via claude -p < capture-thread-417.md from inside the repo you want the changes in.

⬇ Download capture-thread-417.md