sscoble/pi-mono
1
0
Fork 0
mirror of https://github.com/badlogic/pi-mono.git synced 2026-11-05 23:37:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 4

feat(coding-agent): default NVIDIA to Nemotron 3.5 Super VL (#10910)

Browse source 
* feat(coding-agent): default NVIDIA to Nemotron 3.5 Super VL

Co-authored-by: prademacher-nvidia <235594174+prademacher-nvidia@users.noreply.github.com>

* docs(coding-agent): changelog for NVIDIA default model change

* fix(ai): run NVIDIA E2E tests against Nemotron 3.5 Super VL

---------

Co-authored-by: prademacher-nvidia <235594174+prademacher-nvidia@users.noreply.github.com>
This commit is contained in:
Carlos Villela 2026-10-16 13:26:37 -07:00 • committed by GitHub
parent 14ff455a0a
commit 95364a54ef
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 7 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

4
packages/ai/test/stream.test.ts
View file

@ -809,8 +809,8 @@ describe("Generate E2E Tests", () => {
});
});
describe.skipIf(!process.env.NVIDIA_API_KEY)("NVIDIA NIM Provider (Nemotron 3 Ultra via OpenAI Completions)", () => {
const llm = getModel("nvidia", "nvidia/nemotron-3-ultra-550b-a55b");
describe.skipIf(!process.env.NVIDIA_API_KEY)("NVIDIA NIM Provider (Nemotron 3.5 Super VL)", () => {
const llm = getModel("nvidia", "nvidia/nemotron-3.5-super-vl-120b-a12b");
it("should complete basic text generation", { retry: 3 }, async () => {
await basicTextGeneration(llm);

4
packages/coding-agent/CHANGELOG.md
View file

@ -30,6 +30,10 @@
- Changed the `kimi-coding` provider to use OpenAI Chat Completions at `https://api.kimi.com/coding/v1`, the protocol used by the official Kimi Code client. K3, K3-256K, and Kimi For Coding now accept tools added mid-conversation without invalidating the prompt cache. `models.json` overrides of the `kimi-coding` base URL must now include `/v1` ([#10720](https://github.com/earendil-works/pi/issues/10720))
- Changed model switching to drop thinking blocks from earlier assistant messages produced by a different model, instead of sending them as plain text. The new model previously saw the old model's reasoning as part of its replies and could start writing its own reasoning into the visible answer ([#10995](https://github.com/earendil-works/pi/issues/10995))
### Changed
- Changed the default NVIDIA model to `nvidia/nemotron-3.5-super-vl-120b-a12b` ([#10910](https://github.com/earendil-works/pi/pull/10910) by [@cv](https://github.com/cv))
### Fixed
- Pinned internal Pi dependencies to exact release versions to prevent fresh installs from mixing incompatible packages ([#10865](https://github.com/earendil-works/pi/issues/10865))

2
packages/coding-agent/src/core/model-resolver.ts
View file

@ -25,7 +25,7 @@ export const defaultModelPerProvider: Partial<Record<KnownProvider, string>> = {
azure: "gpt-5.4",
"openai-codex": "gpt-6.1-sol",
radius: "balanced",
nvidia: "nvidia/nemotron-3-ultra-550b-a55b",
nvidia: "nvidia/nemotron-3.5-super-vl-120b-a12b",
deepseek: "deepseek-v4-pro",
google: "gemini-3.1-pro-preview",
"google-vertex": "gemini-3.1-pro-preview",