GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
marinarosa
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
marinarosa
Model Tree
marinarosa/minicpm5-1b-vivamais-v1
Input Modalities
Output Modalities
Supported Functionality
MiniCPM5-1B Viva Mais text v4, trained from marinarosa/minicpm5-1b-vivamais-v1 with a
broader Brazilian Portuguese conversational and grounded-Q&A mix.
The training package uses a redacted Viva Mais dashboard Q&A replay set, grounding-negative rows, and filtered Rio 3.1 teacher-distilled rows. The v4 variant intentionally excludes the generic WhatsApp-style bucket and uses a stricter final-answer-only teacher prompt. Teacher rows are filtered to reject verbose wrappers and hidden-reasoning artifacts. No raw WhatsApp exports, client identifiers, identity documents, or local private data are published.