Liquid AI Open d1: d1-3B and d1-omni-600M Decision Models
Melvin Vivas · X post · 2026-10-08 · Open on X
Topics: LLM Fundamentals, Fine-tuning & Model Customization, AI Dev Tools & Productivity · Level: intermediate
Summary
Melvin Vivas says a Colab notebook with examples for Liquid AI's d1-3B decision model is coming. He is building it with his Grok Bot coding agent through Cursor's Cloud agent. The quoted Liquid AI post explains when to use d1-3B (high-quality text and vision decisions) and when to use the experimental d1-omni-600M (when footprint matters). It also links the release blog and the Hugging Face weights.
Key points
- Liquid AI's d1 family is made of 'decision models', with weights you can download, fine-tune and deploy locally.
- d1-3B: for high-quality decisions from text and vision.
- d1-omni-600M: experimental, small model for when footprint matters.
- The weights are on Hugging Face under LiquidAI/d1-3b.
- The creator uses a coding agent (Grok Bot through Cursor Cloud agent) to write example Colab notebooks.
- The Colab notebook had not been released when this was posted.
Resources mentioned
- Open d1: Edge decision models for text, vision, and audio (Liquid AI blog) · article · liquid.ai · free
Liquid AI's blog post announcing the open-weight d1 decision models. - LiquidAI/d1-3B (Hugging Face) · tool · huggingface.co · free
Model card and open weights for the d1-3B text and vision decision model. - d1-omni-600M · tool · huggingface.co · free
Liquid AI's experimental small open-weight multimodal decision model for text with image or audio.
Also in: Try Liquid AI's open d1-3B decision model in Colab (Melvin Vivas on X · notes), Liquid AI releases Open d1 open-weight multimodal decision models (Melvin Vivas on X · notes) - Liquid AI (@liquidai) on X · website · x.com · free
Liquid AI's family of efficient foundation models, used here as the example of a model you can run in the browser with ONNX.
Also in: Trying Liquid AI Decision Models (d1, d1 ultra) via API (Melvin Vivas on X · notes), Liquid RIPL: Low-Latency Inference Engine for LFMs (Melvin Vivas on X · notes), ONNX + Transformers.js: Run Models Like Liquid LFM in the Browser (Melvin Vivas on X · notes), Try Liquid AI's open d1-3B decision model in Colab (Melvin Vivas on X · notes) and 30 more - Grok Bot · tool · x.ai · paid
An AI personal assistant the creator describes as having memory, day planning, task scheduling and the ability to do background work without any coding.
Also in: Turn repeated agent tasks into reusable Skills (Melvin Vivas on X · notes), Using a Grok Bot to find free AI credit giveaways (Melvin Vivas on X · notes), Grok Bot can now use phone calendar, contacts and reminders (Melvin Vivas on X · notes), Grok Bot Adds 1Password Integration for Agent Logins (Melvin Vivas on X · notes) and 129 more - Cursor Cloud agent · tool · cursor.com · paid
Cursor's cloud-hosted coding agent that works on code tasks in the background. - Google Colab · tool · colab.research.google.com · free · recommended by both Bashiri Smith & Melvin Vivas
Free GPU notebooks.
Also in: Run LightOnOCR-3 0.8B and Other Open Models on a Free Colab T4 (Melvin Vivas on X · notes), Try Liquid AI's open d1-3B decision model in Colab (Melvin Vivas on X · notes), Google Colab notebooks for new open models, built by a coding agent (Melvin Vivas on X · notes), Liquid AI's LFM2.5-24B-A2B: a local MoE model for agent tasks (Melvin Vivas on X · notes) and 27 more
Try this
- Read the Open d1 blog post and the d1-3B model card on Hugging Face.
- Look out for the creator's d1-3B Colab notebook when it is released.
- Write a Colab notebook that runs d1-3B on example text and vision decision tasks.
More in LLM Fundamentals
- AI Engineer Vocab Quiz: 11 Acronyms to Know for Interviews
- Colab Notebook for Liquid AI's d1-3B Decision Model
- Testing Liquid AI d1-3B in Colab from a Phone
- Liquid AI releases Open d1 open-weight multimodal decision models
- Claude Haiku 5.5 in Devin: beats Sonnet 5 at about 1/8 the cost
- Liquid AI's open d1 decision models: small multimodal models that run locally