Intelligence

<think>

See also my Large Language Models section ... there you can download the abliterated models I'm using locally(!).

Cellular Automata

I've another page about a nice topic here; w/ (emergent) swarm intelligence; see my Cellular Automata page (TODO).

Large Language Models

These are only my own favorites.. there are others, too; but I don't really like using public versions.. only sometimes.

Gemini Google ChatGPT OpenAI

Inference

These three are the apps I've already tested before.

LM Studio llama.cpp ollama

These ones were not tested by me.. I've only heard about them. Maybe good?

Jan vllm msty mlx-lm

Models

I really like running LLMs on my local machine, or smth. not accessable by others. Even if I'm pretty sure, E.T./A.I. phone home... xD~

Open LLM Leaderboard

Here are my favorite models (currently) ... sorted by ~~date~~/...randomly! ...

My Tipp: See also the abliterated models section below (with download link)! ...

cwm Qwen3-Next-80B-A3B-Thinking-FP8 Qwen3-Next-80B-A3B-Thinking ERNIE-4.5-21B-A3B-Thinking GLM-4.5 DeepSeek-V3.2-Exp DeepSeek-R1-0528 Qwen3-Coder-480B-A35B-Instruct Qwen3-235B-A22B-Thinking-2507 MiniMax-M1-80k MiniMax-Text-01 Kimi-K2-Base Devstral-Small-2507 Mistral-Large-Instruct-2411 medgemma-27b-text-it Mixtral-8x22B-v0.1 DeepSeek-Coder-V2-Instruct-0724 Llama-4-Maverick-17B-128E gemma-3-27b-it Codestral-22B-v0.1 Qwen3-8B phi-4 phi-4-reasoning-plus Qwen3-Coder-30B-A3B-Instruct Qwen3-32B Magistral-Small-2509 Qwen3-30B-A3B-Instruct-2507 Devstral-Small-2507 Llama-3.3-70B-Instruct Qwen3-Coder-480B-A35B-Instruct Mistral-Large-Instruct-2411 CodeLlama-34b-Instruct-hf Mistral-Small-3,2-24B-Instruct-2506 gpt-oss-120b

Abliterated Models

Here are some abliterated models @ Hugging Face (please explicitly search for 'em there).

JFYI: Now you can get a copy of all my abliterated models here!

failspy huihui-ai mlabonne mradermacher

It's about uncensoring LLMs ... the "easy" way (in a "global" form, not as usual by changing the prompt(s)). Find out more here:

Uncensor any LLM with abliteration #1 Uncensor any LLM with abliteration #2 Refusal in LLMs is mediated by a single direction #1 Refusal in LLMs is mediated by a single direction #2 Demo of bypassing refusal LLM Abliterate v1.4 script, adapted for Mistral-Small-Instruct-2409

Tiny helper Scripts

Created by myself, to make it easier for me to handle the models, etc..

convert-hf-to-gguf.sh hfdownloader.sh hfget.sh

Links / Resources

I collected the interesting links (for myself) while surfing the web..
They aren't sorted in any priority or smth. like that ... they're just sorted by date!

Hugging Face (!) Large Language Model Course Neural Network Zoo "What are the risks from Artificial Intelligence?" "Machine learning in a few clicks" AutoDAN-Turbo: A Lifelong Agent for Strategy Self-Exploration to Jailbreak LLMs Genesis LLM Visualization Spreadsheets are all you need NodeJS library for Llama Stack The GPT-3 Architecture, on a Napkin Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs Microsoft KBLaM: Knowledge Base augmented Language Model MICrONS Explorer: A virtual observatory of the cortex CL4R1T4S (System Prompt Transparency for all) Echo Chamber: A Context-Poisoning Jailbreak That Bypasses LLM Guardrails How Can AI ID a Cat? An Illustrated Guide. Bias in der künstlichen Intelligenz A Visual Guide to Quantization A Gentle Introduction to 8-bit Matrix Multiplication for transformers at scale using Hugging Face Transformers, Accelerate and bitsandbytes Binary and Scalar Embedding Quantization for Significantly Faster & Cheaper Retrieval The Illustrated GPT-2 (Visualizing Transformer Language Models) The Illustrated DeepSeek-R1

Continuous Thought Machines

I don't know how this is scaling in real, but the idea itself seems interesting!
See (w/ it's own GitHub repository).

Neural Network Zoo

The official link is above, in the list of Links. I don't know all these types of neural networks, but it looks like an interesting overview.

I'm also hosting the image itself here. So here it comes (and, btw., click on it to open the full view in a new browser tab):

`Norbert`

This is my very own artificial intelligence.

It's totally based on pure byte code, nothing with tokens or smth. like it (like current LLMs are based on).
And such input is processed abstract. The last output is then also injected once again as second, parallel input byte (some feedback loop).

Screenshots

And here are some screenshots of the process itself and two helper utilities I've created especially for this reason; also my dump.js.