LibertAI Labs
Latest note — Uncensoring GLM-5.3-Flash: one agent, two DGX Sparks, one night

Where boundaries are made to be pushed.

The experimental research branch of LibertAI — open weights, novel inference architectures and agent systems for the next generation of private, decentralized intelligence.

24k+
9
9

Model releases

Open weights, quantization notes and reproducible benchmark context.

Agent systems

Runtime experiments, evaluation harnesses and tool interfaces for private agents.

Infrastructure

Search, confidential compute and payment primitives that support the product stack.

Releases and research tracks

Shipped

NVFP4 model releases

GGUF and safetensors builds of flagship open models with calibrated NVFP4 tensors, benchmarked honestly against stock K-quants on Blackwell hardware.

Read project →
Shipped

Consensus web search

A search API that queries Google, Bing and DuckDuckGo in parallel from Aleph Cloud's decentralized network and ranks results by cross-engine agreement.

Read project →
Shipped

LibertAI Code

A coding agent for the terminal and desktop that runs open models on confidential, TEE-backed inference, with no training on your code.

Read project →
Experiment

Conscio

The consciousness layer for LLM agents: persistent memory, attention, drives, self-monitoring, reflection, and autonomous action in one inspectable runtime.

Read project →
Shipped

MCP server

A Model Context Protocol server built into the CLI that gives Claude Code, Cursor, Cline and any other MCP client access to consensus web search.

Read project →
Experiment

Avatar

A browser avatar you can talk to, combining a 3D stage, browser speech APIs and LibertAI chat completions, with conversations kept in your browser.

Read project →
Shipped

GLM-5.3-Flash on vLLM, on GB10

A hand-written sparse-MLA CUDA kernel for sm_121 and a fix for vLLM's uninitialised NVFP4 MoE activation scale, which together take GLM-5.3-Flash from degenerate output to correct serving on a two-node DGX-Spark-class cluster.

Read project →
Research

Post-quantum blockchain security review

An adversarial review of 30 live networks, migration efforts, standards proposals and on-chain verifiers, with high-priority findings moving through responsible disclosure.

Read project →

Published on Hugging Face

Quantized builds of the open models we serve, published under LibertAIDAI with benchmark numbers in every model card. Counts are from June 2026.

View on Hugging Face ↗
repos
9
downloads
24k+
base models
5

Formal writeups

Full papers with related work, falsifications and precise claims.

Methodology and results

Writeups with methodology, the numbers that held up, and the ones that did not.