Liquid AI ORG

Examples, end-2-end tutorials and apps built using Liquid AI Foundational Models (LFM) and the LEAP SDK

Liquid Audio - Speech-to-Speech audio models by Liquid AI

No description.

A full model customization repo for Liquid Foundation Models

Example apps for LeapSDK

No description.

No description.

No description.

Liquid documentation

No description.

No description.

No description.

Scripts to launch Liquid on-prem stack. This repo is deprecated, since LFMs are now open source.

Python example to run LFM2 with llama.cpp

Reproducible encoder downstream fine-tune eval harness

LLM inference in C/C++

No description.

No description.

LEAP Multiplatform SDK

Official samples for Liquid LFMs on Amazon Bedrock

Modified mt_bench with API and HF scripts for LFMs.

Faster structured generation

No description.

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent

Modified llm-jp-eval with API and HF scripts for LFMs.

LLM inference in C/C++

Modified RULER with API scripts for LFMs.

Open-source evaluation toolkit of large multi-modality models (LMMs), support 220+ LMMs, 80+ benchmarks

Gorilla: Training and Evaluating LLMs for Function Calls (Tool Calls)

Rigourous evaluation of LLM-synthesized code - NeurIPS 2023 & COLM 2024

Minimalist ML framework for Rust

Gorilla: Training and Evaluating LLMs for Function Calls (Tool Calls)

A high-throughput and memory-efficient inference and serving engine for LLMs

No description.

τ²-Bench: Evaluating Conversational Agents in a Dual-Control Environment

Batteries-included eval framework for search APIs

C# example to run LFM2 with llama.cpp / LlamaSharp

Python bindings for llama.cpp

Provider-agnostic, open-source evaluation infrastructure for language models

On-device AI privacy proxy — releases

Diagram builder in the Tidal Design System style — Tidal-styled canvas, Mermaid & image import, local library

Gorilla: Training and Evaluating LLMs for Function Calls (Tool Calls)

Welcome to GR00T Whole-Body Control (WBC)! This is a unified platform for developing and deploying advanced humanoid controllers. This includes: Decoupled WBC models used in NVIDIA Isaac-Gr00t, Gr00t N1.5 and N1.6 and GEAR-SONIC

Dataflow-Oriented Reinforcement Learning for (Multi-)Agentic LLMs

LLMs and VLMs with MLX Swift

No description.