Alessandro Ercolani PRO
giux78
AI & ML interests
NLP, Reinforcement Learning, Semantics, Computational Neuroscience
Recent Activity
reacted
to
their post with 🔥 about 10 hours ago
Together with @mferraretto and @efederici we released #Nesso-4B, a new model specialized for agentic workflows.
https://huggingface.co/mii-llm/nesso-4B
#Nesso-4B is a fine-tuned version of Qwen-4B, trained on a highly curated and balanced dataset designed specifically for multilingual agentic workflows and conversational use cases.
As shown in the video below we simulate, the new “cowork” from #Antrophic, without any data sharing all running on a consumer device. The model can be used to build agentic behavior in #privateAI environments.
Not every problem requires super intelligence: in many cases, intelligence at the edge is more than enough.
#Nesso4B #AgenticAI #PrivateAI #EdgeAI #OnDeviceAI reacted
to
robtacconelli's
post with 🚀 about 10 hours ago
🏆 Nacrith: a 135M model that out-compresses everything on natural language
What if a tiny LM could compress english text better than _every_ compressor out there — classical or neural, small or large?
Nacrith pairs SmolLM2-135M with an ensemble of online predictors and high-precision arithmetic coding.
What's inside
The standard LLM+arithmetic coding approach wastes ~75% of CDF precision on large vocabularies. Our CDF-24 fix alone recovers 0.5 bpb. On top: a token N-gram that skips the GPU on predictable tokens, an adaptive bias head, llama.cpp backend (7× faster than PyTorch), multi-GPU parallel compression, and a binary file format (NC06) — the first LLM-based binary compressor we know of.
Runs on a GTX 1050 Ti. ~500 MB weights, ~1.2 GB VRAM per worker.
💻 Code: https://github.com/robtacconelli/Nacrith-GPU
⭐ Space: https://huggingface.co/spaces/robtacconelli/Nacrith-GPU
📄 Paper: https://huggingface.co/papers/2602.19626
Try it, break it, share your results — all feedback welcome. ⭐ on the repo appreciated!
Results across all systems we tested:
- alice29.txt → 0.918 bpb (−44% vs CMIX, −20% vs ts_zip) — below the 2nd-order Shannon entropy bound
- enwik8 (100 MB) → 0.9389 bpb (−8% vs FineZip/LLMZip's 8B model, −15% vs ts_zip)
- Unseen text → 0.723 bpb on a doc published after training cutoff — no memorization, 26% better than FineZip/LLMZip on the same model
SmolLM2-135M by https://huggingface.co/HuggingFaceTB liked
a model 21 days ago
mii-llm/nesso-4B Organizations
The leaderboard is down...
2
#1 opened over 1 year ago
by
zhiminy
Update app.py
1
#12 opened almost 2 years ago
by
giux78
Access Problems
👍 👀 12
61
#45 opened almost 2 years ago
by
VityaVitalich
Dataset is not loading
2
#2 opened almost 2 years ago
by
vinbloke
Information on the model
4
#1 opened almost 2 years ago
by
anakin87
What is `m_mmul` benchmark?
3
#7 opened almost 2 years ago
by
zhiminy
Upload folder using huggingface_hub
2
#1 opened almost 2 years ago
by
giux78
Upload folder using huggingface_hub
2
#1 opened almost 2 years ago
by
giux78