You can deploy this model and start earning money today!

GurkeBaui/
Karla

Name: Karla
Availability: InStock

interstellarninja/hermes_reasoning_tool_use

crownelius/Opus-4.6-Reasoning-2100x-formatted

ronantakizawa/github-top-code

2512.24695

2505.05522

nvidia/Cosmos-Reason2-2B

finetune

mit

Views

No views yet

Karla C1: The Adaptive Brain for Physical AI

Karla C1 is a novel, continuous-learning architecture designed for Physical AI and embodied agents. Built on top of NVIDIA Cosmos-Reason2-2B, it solves the problem of catastrophic forgetting in robotics by implementing a Nested Learning Architecture [1] with Surprise-Based Plasticity.

The Vision: Physical AI on the Edge

Robots in the real world encounter new tools and constraints daily. Karla allows an agent to learn a new physical tool or API on the fly during inference, modifying its own weights in seconds on consumer hardware (e.g., RTX 4060 Ti), without ever forgetting its foundational knowledge.

Architecture: Nested Learning

Instead of a standard transformer pipeline, Karla acts as a multi-frequency brain based on the Nested Learning (NL) paradigm [1]:

L0 (The Subconscious): NVIDIA Cosmos-Reason2-2B (Frozen). Provides world-class reasoning, semantic understanding, and syntax. Loaded in 4-bit to save VRAM.
L1 (The RAM): Dynamic Knowledge MoE (64 Experts). A fast, GPU-based Mixture of Experts. It updates live during inference using Delta Gradient Descent [1], memorizing new facts and tool syntaxes instantly.
L2 (The Frontal Lobe): Continuous Thought Machine (CTM). Based on Sakana AI's research [2], this is a parallel sequence-level reasoning module. It takes Cosmos's hidden states + L1 knowledge and "thinks" for $T$ internal ticks before outputting an action plan.

[!WARNING] The L2 module currently faces stability issues where the CTM's internal dynamics can lead to token repetition during long reasoning chains.

Quickstart

bash

pip install torch transformers datasets pandas accelerate bitsandbytes
cd Karla
python chat.py

TODO

Train Longer
Make L1 Inference Learning Intensity variable
Give the model control over itself
Fix stuttering

Datasets Used

interstellarninja/hermes_reasoning_tool_use
crownelius/Opus-4.6-Reasoning-2100x-formatted
ronantakizawa/github-top-code

References

[1] Behrouz, A., Razaviyayn, M., Zhong, P., & Mirrokni, V. (2025). Nested Learning: The Illusion of Deep Learning Architecture. Google Research. arXiv:2512.24695

[2] Darlow, L., Regan, C., Risi, S., Seely, J., & Jones, L. (2025). Continuous Thought Machines. Sakana AI. arXiv:2505.05522

GurkeBaui/Karla