Jev-style decision models fine-tuned to judge cheap-model code
James PRO
jtatman
AI & ML interests
improving domain specific models and re-sampling data, refining datasets for use in different modalities, small scale micro-llm clusters using quantized and smoothed models, and all emerging llm stack connecting technologies. Small models rock.
Recent Activity
liked a dataset 10 days ago
XiaomiMiMo/MiMo-V2.6-RL-oss liked a model 12 days ago
Qwen/Qwen3.5-0.8B upvoted an article 13 days ago
Building the Open Agent Ecosystem Together: Introducing OpenEnvOrganizations
Small Models Fine-Tuned for Action
-
jtatman/Reyna-Mini-1.8B-v0.2-function-call-laser
Text Generation • 2B • Updated • 43 • • 7 -
jtatman/TinyDolphin-3x-MoE
Text Generation • 3B • Updated • 128 • 1 -
jtatman/Reyna-Mini-1.8B-v0.2-cosmopedia-laser
Text Generation • 2B • Updated • 13 -
jtatman/tinydolphin-3x-MoE-samantha-chatml-therabot
Text Generation • 3B • Updated • 20 • 1
Mix and Merge
Jev-style decision models fine-tuned to judge cheap-model code
Small Models Fine-Tuned for Action
-
jtatman/Reyna-Mini-1.8B-v0.2-function-call-laser
Text Generation • 2B • Updated • 43 • • 7 -
jtatman/TinyDolphin-3x-MoE
Text Generation • 3B • Updated • 128 • 1 -
jtatman/Reyna-Mini-1.8B-v0.2-cosmopedia-laser
Text Generation • 2B • Updated • 13 -
jtatman/tinydolphin-3x-MoE-samantha-chatml-therabot
Text Generation • 3B • Updated • 20 • 1
quantized models that will run on almost anything
Mix and Merge