1. dont fall for anti open model fearmongering, but 2. acknowledge that AI capabilities are proceeding fast, and eventually there may be a reason to be more careful with open weight models I don't think Mythos is that trigger, but I'm not 100% confident www.interconnects.ai/p/claude-myt...
Claude Mythos and misguided open-weight fearmongering
Another dance around fears of open-source.
interconnects.ai
made Claude Opus out of GPT-5.4. well, sort of. I used a genetic algorithm to make GPT act more like Claude on coding tasks. Not by making it smarter. Mostly by tuning the boring stuff that changes how it feels in practice: tool cadence, stop/go judgment, how deep it digs, and when it stops.
Created ngi inspired from the latest Cursor post. It is a faster and more efficient alternative to grep for agents. Already using it and works nicely! github.com/erogol/ngi
i was vibing on a few LLM projects with zero visibility into what was actually happening. silent retry bugs burning tokens, wrong API keys, traffic to dead endpoints. couldn't debug any of it. so i created this github.com/erogol/toklog
GitHub - erogol/toklog: htop for your LLM endpoints.
htop for your LLM endpoints. Contribute to erogol/toklog development by creating an account on GitHub.
github.com
Agentic tools like OpenClaw grow with every PR — but bigger codebases are harder for AI to understand and extend. What if we kept the core tiny and let agents adapt themselves to user needs? No PR, just evolve.
My post on MiMo-Audio open.substack.com/pub/erogol/p... 🔥 Trained on 100M+ hours and shows emergent few-shot learning: • Voice conversion • Emotion transfer• Speech translation • Cross-modal reasoning ⚡ Key finding: Speech follows same scaling laws as text LLMs
Model Check - MiMo-Audio: Scaling Speech Pre-Training to 100M Hours
Going over the code and the technical report of the new Speech LM model from Xiaomi that rivals GPT4o-audio and Gemini
open.substack.com
Machine Learns #55 is out! Full of new models… check it out open.substack.com/pub/erogol/p...
Machine Learns #55
Voice + reasoning releases (Ling‑flash‑2.0, VoxCPM, Kimi K2, ultraVAD) and 2 papers: long‑horizon execution & decay‑free LR schedules.
open.substack.com
My breakdown of VibeVoice - new open-weight TTS model from Microsoft. open.substack.com/pub/erogol/p...
Model Check - VibeVoice: Next-Token Diffusion Meets Long-Form Speech Generation
Going over the code and the technical report of the new TTS model from Microsoft Research.
open.substack.com
ms released a tts model… nice… You can create long form convos and podcasts with 4 distinct voice huggingface.co/microsoft/Vi...
microsoft/VibeVoice-1.5B · Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
KyutaiTTS solved streaming text-to-speech with a state machine that generates audio word-by-word as text arrives. 220ms latency, 10-second voice cloning, 32 concurrent users on single GPU. No more waiting for complete sentences. Full analysis: erogol.substack.com/p/model-chec...
Model check - KyutaiTTS: Streaming Text-to-Speech with Delayed Streams Modeling
Going over the Kyutai's new TTS model and its delayed streaming model.
erogol.substack.com
This is such a great idea
Sakana AI@sakanaai.bsky.social · last yr.We’re excited to introduce Text-to-LoRA: a Hypernetwork that generates task-specific LLM adapters (LoRAs) based on a text description of the task. Catch our presentation at #ICML2025! Paper: arxiv.org/abs/2506.06105 Code: github.com/SakanaAI/Tex...
claude is the best coding model gemini cause frequent syntax errors openai does not even understand the task at hand
lately spending sometime with Diffusion LMs and working on NanoGPT style LlaDA model so far I've not achieved comparable results to AR models but its a good start github.com/erogol/BlaGP...
BlaGPT/bla_gpt/llada.py at main · erogol/BlaGPT
Experimental playground for benchmarking language model (LM) architectures, layers, and tricks on smaller datasets. Designed for flexible experimentation and exploration. - erogol/BlaGPT
github.com
This work was done in collaboration with Jeff Clune’s lab at UBC, and led by his PhD students Jenny Zhang and Shengran Hu, together with Cong Lu and Robert Lange. Paper: arxiv.org/abs/2505.22954 Code: github.com/jennyzzt/dgm
⚡ Machine Learns issue 48 is out 🚀 dKV-Cache accelerates diffusion models up to 10x faster 🔐 OpenAI's authentication play (think OAuth for AI) 🎯 PaTH Attention beats RoPE on long-context tasks 🤖 Humanoid Robot fights became real open.substack.com/pub/erogol/p...
Machine Learns #48
OpenAI's 'Sign in with ChatGPT', Meta's AGI ambitions, new models like Gemma 3 & MAGI-1, research breakthroughs in KV caching for diffusion & PaTH Attention, and fresh open-source releases.
open.substack.com
Following the bread crumbs, implemented PLE from Gemma3n. It gave a significant performance boost and resulted in a new best model with almost no compute overhead. github.com/erogol/BlaGPT
GitHub - erogol/BlaGPT: Experimental playground for benchmarking language model (LM) architectures, layers, and tricks on smaller datasets. Designed for flexible experimentation and exploration.
Experimental playground for benchmarking language model (LM) architectures, layers, and tricks on smaller datasets. Designed for flexible experimentation and exploration. - erogol/BlaGPT
github.com
My paper notes on 2 new papers - Model Merging in Pre-training of Large Language Models, - Do Not Let Low-Probability Tokens Over-Dominate in RL, open.substack.com/pub/erogol/p...
Paper check: Merging LLMs at Pre-training, Considering Token Probabilities at RL
🔬Two papers in scope: "Model Merging in Pre-training for LLMs" and "Do Not Let Low-Probability Tokens Over-Dominate in RL"
open.substack.com
muon really works. got best results in BlaGPT ``` torchrun --standalone --nproc_per_node=8 train.py --run_name best_model --model_name best ``` github.com/erogol/BlaGPT
GitHub - erogol/BlaGPT: Experimental playground for benchmarking language model (LM) architectures, layers, and tricks on smaller datasets. Designed for flexible experimentation and exploration.
Experimental playground for benchmarking language model (LM) architectures, layers, and tricks on smaller datasets. Designed for flexible experimentation and exploration. - erogol/BlaGPT
github.com
🧵 Here is a small thread with my notes about some of the recent Transformer papers. - Softpick: an alternative to softmax in Attention - Canon Layers: mixing states with conv1d - Parallel Transformer blocks
Machine learns #45 - no fluff AI newsletter - is out! I normally share bi-weekly but last week was full enough so here we go open.substack.com/pub/erogol/p...
Machine Learns #45
OpenAI's social network & GPT-4.1, China launches $8.2B AI fund, NVIDIA's US manufacturing push, new GLM-4 & MineWorld models, C3PO expert pathways optimization, GigaTok's 3B visual tokenizer...
open.substack.com
Updated my LLM usage and cancelled ChatGPT sub for now Coding - Claude, Gemini 2.5 Reading papers - Claude Research - Gemini 2.5 Daily - Gemini 2.5 Search - Gemini 2.5
@erogol.com · last yr.Here is my use of LLMs Coding - Claude (best by far), QwenChat Reading papers - Claude Research - ChatGPT (best UI,UX), Gemini (better results) Daily - ChatGPT Search - ChatGPT I'd love to try searching with Claude, but not there yet. Any suggestions for change?
Machine Learns #44 is out !! click for no fluff AI newsletter erogol.substack.com/p/machine-le...
Machine Learns #44
Praxis Sam Altman's tech utopia, Amazon launches Nova Sonic voice AI, Midjourney returns with V7, Llama 4 models debut amid controversy, new brain-to-voice model, NoProp learning ...
erogol.substack.com
Next big thing is Brain-LLMs. Imagine an LLM compressing all world knowledge attached to your brain and ready to serve your thoughts and questions. You also update it over internet and pay for sub. I don't want to think about the ad business :)
“If these results generalize to real-world software tasks, extrapolation of this trend predicts that within 5 years, AI systems will be capable of automating many software tasks that currently take humans a month.” arxiv.org/abs/2503.14499
Measuring AI Ability to Complete Long Tasks
Despite rapid progress on AI benchmarks, the real-world meaning of benchmark performance remains unclear. To quantify the capabilities of AI systems in terms of human capabilities, we propose a new me...
arxiv.org
Here is my use of LLMs Coding - Claude (best by far), QwenChat Reading papers - Claude Research - ChatGPT (best UI,UX), Gemini (better results) Daily - ChatGPT Search - ChatGPT I'd love to try searching with Claude, but not there yet. Any suggestions for change?
I think diffusion-based LLMs (LLdMs) are better suited as next-generation LLMs - multiple outputs per iter: faster output generation - no causal masking: bidirectional attention - multiple diff steps: reasoning at inference time and revising poor outputs
