gianlucamazza/SmolLM2-360M-Instruct-ort-genai-int4-cpu
Text Generation • Updated
Gianluca Mazza PRO
gianlucamazza
AI & ML interests
LLM systems in production, agent state and recovery, evaluation, inference cost
Recent Activity
posted an update about 4 hours ago
A live demo of emotional-memory 0.18.1: chat, watch the affective state move, and type "recall X" to see memories ranked by meaning plus mood. It shows how the mechanism works, not that it wins in general. The Scope and limits section lists the benchmarks where plain cosine beats it.
https://huggingface.co/spaces/gianlucamazza/emotional-memory-demo reacted to alibidaran's post with 🧠 1 day ago
I have recently worked on a personal project called progressive thinking, which enables LLMs to generate chains of thought in a progressive pattern. In addition, the model learned to generate COT before choosing the appropriate tools, making it useful for complex and deeper agentic tasks. This model is well trained on QWEN3-14B using the SFT algorithm and needs GRPO to make feasible outputs.
https://huggingface.co/alibidaran/qwen_progressive_thinking_16bit
https://huggingface.co/datasets/OpenDataArena/Spark-234K
https://huggingface.co/datasets/IFM/TxT360-3efforts posted an update 1 day ago
I've reposted my write-up on running a small language model on an Xbox Series S. The open problems come first, and the later updates that corrected some of my own numbers are kept in, near the end. The original is at gianlucamazza.it/en/blog/slm-on-xbox.
https://huggingface.co/blog/gianlucamazza/running-a-small-language-model-on-an-xbox-series-sOrganizations
