Language models trained from scratch on a single consumer GPU.
De Francesco Nazareno
nazdef
·
AI & ML interests
LLM pretraining from scratch, single-GPU training, bilingual EN/IT models, Supervised finetuning SFT, Reinforcement learning GRPO/GSPO
Recent Activity
updated a collection 24 days ago
1GPU LLM updated a collection 25 days ago
1GPU LLM updated a collection about 1 month ago
1GPU LLMOrganizations
None yet
models 40
nazdef/1gpu-llm-small-en-it-base
Text Generation • 0.1B • Updated • 21
nazdef/1gpu-llm-small-en-it-base-v2
Text Generation • 0.1B • Updated • 16
nazdef/1gpu-llm-medium-en-it-base-v2
Text Generation • 0.3B • Updated • 8
nazdef/20260713_resume-gpt2medium-step31000-d2000-step33000
Text Generation • 0.3B • Updated • 6
nazdef/20260713_resume-gpt2medium-step22000-d1800-step23100
Text Generation • 0.3B • Updated • 7
nazdef/20260703_continual-pretraining-gpt2medium-step34000-nodecay
Text Generation • 0.3B • Updated • 6
nazdef/1gpu-llm-medium-en-it-base
Text Generation • 0.3B • Updated • 8
nazdef/gpt2small-en-it-nanochat-gpt2preln-cpt8600-lr5e5-predecay-step18000
Text Generation • 0.1B • Updated • 16
nazdef/gpt2medium-en-it-nanochat-gpt2preln-decay13500-step14700
Text Generation • 0.3B • Updated • 15
nazdef/gpt2medium-en-it-nanochat-gpt2preln-decay13500-step14500
Text Generation • 0.3B • Updated • 17
datasets 0
None public yet