The Troiani LLM model family progression from v0 -> v1 including Base, Instruct and specializations like Vision, Audio...
Walter Troiani Vargas
eZWALT
·
AI & ML interests
None yet
Recent Activity
liked a model 22 days ago
anarlavrenov/lime-1b-instruct updated a collection about 1 month ago
Troiani v1 published a dataset about 1 month ago
eZWALT/Instruct-SFT-Troiani-v0Organizations
Production LLMs
- RunningFeatured1.4k
FineWeb: decanting the web for the finest text data at scale
🍷1.4kExplore and download the FineWeb web‑scale text dataset
- Running3.96k
The Ultra-Scale Playbook
🌌3.96kThe ultimate guide to training LLM on large GPU Clusters
- RunningAgents114
Predict Memory
🧮114Estimate model memory usage and see detailed plots
- Running50
Encoder-Free VLM
👁50Train Your Own Encoder-Free VLM in $100
Multimodal NanoChimera
-
HuggingFaceTB/SmolVLM-Instruct
Image-Text-to-Text • 2B • Updated • 24.5k • 592 -
google/siglip2-base-patch16-224
Zero-Shot Image Classification • 0.4B • Updated • 785k • 121 -
openai/clip-vit-base-patch32
Zero-Shot Image Classification • Updated • 22.9M • 992 -
google/siglip2-base-patch16-512
Zero-Shot Image Classification • 0.4B • Updated • 78.4k • 49
Cursed Toxic Pretraining Corpora
Troiani v1
The Troiani LLM model family progression from v0 -> v1 including Base, Instruct and specializations like Vision, Audio...
Production LLMs
- RunningFeatured1.4k
FineWeb: decanting the web for the finest text data at scale
🍷1.4kExplore and download the FineWeb web‑scale text dataset
- Running3.96k
The Ultra-Scale Playbook
🌌3.96kThe ultimate guide to training LLM on large GPU Clusters
- RunningAgents114
Predict Memory
🧮114Estimate model memory usage and see detailed plots
- Running50
Encoder-Free VLM
👁50Train Your Own Encoder-Free VLM in $100
Master Thesis: LLMs <> Ads
Multimodal NanoChimera
-
HuggingFaceTB/SmolVLM-Instruct
Image-Text-to-Text • 2B • Updated • 24.5k • 592 -
google/siglip2-base-patch16-224
Zero-Shot Image Classification • 0.4B • Updated • 785k • 121 -
openai/clip-vit-base-patch32
Zero-Shot Image Classification • Updated • 22.9M • 992 -
google/siglip2-base-patch16-512
Zero-Shot Image Classification • 0.4B • Updated • 78.4k • 49
RLHF Resources
Cursed Toxic Pretraining Corpora