Update README.md

README.md

@@ -9,11 +9,11 @@ pinned: false
 
 <div align="center">
 
-# GestaltLabs
+# Gestalt Lab 🇨🇦
 
-**Open model research for reasoning, agents, and local-first multimodal AI.**
+**Independent Canadian Open Reasoning Research**
 
-GestaltLabs is the research home for the Ornstein, Harmonic, Acta, Talos, and related releases by [DJLougen](https://huggingface.co/DJLougen). We build, curate, and publish models and datasets for practical reasoning systems: multimodal assistants, tool-using agents, local GGUF deployments, and post-training experiments around capability preservation, refusal ablation, and high-signal synthetic data.
+GestaltLabs is the research home for the **Ornstein**, **Harmonic**, **CHIMERA**, **Acta**, **Talos**, and related model lines by [DJLougen](https://huggingface.co/DJLougen). We build, curate, and release **fully permissive open weights** for practical reasoning systems: multimodal assistants, tool-using agents, local GGUF deployments, and post-training experiments around capability preservation, refusal ablation, and high-signal synthetic data.
 
 [Models](https://huggingface.co/GestaltLabs?sort_models=lastModified) · [Datasets](https://huggingface.co/datasets?author=DJLougen) · [DJLougen](https://huggingface.co/DJLougen)
 
@@ -21,19 +21,24 @@ GestaltLabs is the research home for the Ornstein, Harmonic, Acta, Talos, and re
 
 ---
 
-## Current Focus
+## Why the Flag?
+
+**Canadian Open Weights.**  
+Released independently and permissively — no contact forms, no non-commercial clauses, no gatekeeping. Real sovereignty means weights you can actually download, fine-tune, quantize, and deploy without asking permission.
+
+---
 
-- **Ornstein** - multimodal and MoE model releases for reasoning, conversation, and image-text workflows
-- **CHIMERA** - reasoning-oriented models and datasets for self-correction, chain-of-thought style supervision, and agent traces
-- **SABER / RYS** - post-training experiments around refusal ablation, capability preservation, layer surgery, and model behavior editing
-- **Acta / Talos** - curated agentic tool-use and coding-assistant traces for SFT and evaluation
-- **Local inference** - GGUF, quantized, and deployment-friendly builds for llama.cpp-compatible workflows
+## Current Focus
+- **Ornstein** — multimodal and MoE reasoning models (Qwen/Gemma-derived) optimized for local deployment and agentic workflows
+- **CHIMERA** — reasoning-oriented models and datasets emphasizing self-correction, chain-of-thought supervision, and agent traces
+- **SABER / RYS** — post-training experiments around capability preservation, refusal ablation, and model behavior editing
+- **Acta / Talos** — curated agentic tool-use and coding-assistant traces for SFT and evaluation
+- **Local inference** — GGUF, quantized, and deployment-friendly builds for llama.cpp and MLX workflows
 
 ## Featured Model Lines
 
 ### Ornstein
-
-The Ornstein line explores multimodal reasoning and local deployment across Qwen and Gemma-derived models, including 27B, 31B, and 35B-A3B variants.
+Multimodal reasoning line across 27B–35B scales with strong local deployment characteristics.
 
 - [GestaltLabs/Ornstein-3.6-27B](https://huggingface.co/GestaltLabs/Ornstein-3.6-27B)
 - [GestaltLabs/Ornstein-3.6-27B-GGUF](https://huggingface.co/GestaltLabs/Ornstein-3.6-27B-GGUF)
@@ -41,32 +46,27 @@ The Ornstein line explores multimodal reasoning and local deployment across Qwen
 - [DJLougen/Ornstein-27B-v2-GGUF](https://huggingface.co/DJLougen/Ornstein-27B-v2-GGUF)
 
 ### CHIMERA
-
-coming 
+*Coming soon* — reasoning-focused models with explicit self-correction and agent trace supervision.
 
 ### SABER / RYS
-
-SABER and RYS are experimental post-training branches for studying model behavior, refusal boundaries, and capability-preserving edits.
+Experimental post-training branches studying refusal boundaries and capability-preserving edits.
 
 - [DJLougen/gemma-4-E2B-it-saber](https://huggingface.co/DJLougen/gemma-4-E2B-it-saber)
 
 ## Featured Datasets
-
-- [Ornstein Curated 100K](https://huggingface.co/datasets/DJLougen/ornstein-curated-100k) - curriculum-sorted multi-domain reasoning data
-- [Hermes Agent Traces Filtered](https://huggingface.co/datasets/DJLougen/hermes-agent-traces-filtered) - quality-filtered agent reasoning traces
-- [Harmonic Reasoning v1](https://huggingface.co/datasets/DJLougen/harmonic-reasoning-v1) - compact synthetic reasoning data for math, code, and self-correction
-- [WittgenSite](https://huggingface.co/datasets/DJLougen/wittgensite) - prompt consistency benchmark for AI coding agents
-- [Acta](https://huggingface.co/datasets/DJLougen/Acta) - curated agentic tool-use conversations
-- [Talos Scenarios](https://huggingface.co/datasets/DJLougen/Talos-Scenarios) - agentic task scenarios for synthetic trace generation
+- [Ornstein Curated 100K](https://huggingface.co/datasets/DJLougen/ornstein-curated-100k) — curriculum-sorted multi-domain reasoning data
+- [Hermes Agent Traces Filtered](https://huggingface.co/datasets/DJLougen/hermes-agent-traces-filtered) — quality-filtered agent reasoning traces
+- [Harmonic Reasoning v1](https://huggingface.co/datasets/DJLougen/harmonic-reasoning-v1) — compact synthetic reasoning data for math, code, and self-correction
+- [WittgenSite](https://huggingface.co/datasets/DJLougen/wittgensite) — prompt consistency benchmark for AI coding agents
+- [Acta](https://huggingface.co/datasets/DJLougen/Acta) — curated agentic tool-use conversations
+- [Talos Scenarios](https://huggingface.co/datasets/DJLougen/Talos-Scenarios) — agentic task scenarios for synthetic trace generation
 
 ## Principles
-
-- Ship open artifacts that people can inspect, run, and adapt
-- Pair model releases with practical local inference formats
-- Treat datasets as research objects, not just training fuel
-- Explore behavior editing while preserving useful capabilities
-- Build for researchers, tinkerers, and builders who want systems they can actually run
+- Ship **open artifacts** that people can inspect, run, and adapt without restrictions
+- Pair every model release with practical local inference formats (GGUF, MLX, etc.)
+- Treat datasets as first-class research objects, not just training fuel
+- Explore behavior editing (refusal ablation, layer surgery) while preserving useful capabilities
+- Build for researchers, tinkerers, and builders who want systems they can actually run locally
 
 ## Contact
-
-For questions, collaboration, eval results, or licensing discussions, open a discussion on the relevant repository or reach out through [DJLougen](https://huggingface.co/DJLougen).
+Open a discussion on any repository or reach out via [DJLougen](https://huggingface.co/DJLougen).