A family of System One-style models fine-tuned from Qwen3.5, designed for one-pass typed decisions with calibrated probabilities.
Mark Marosi
Mapika
P(doom) 1%
AI & ML interests
NLP, Drug Discovery
Recent Activity
updated a dataset 7 days ago
Mapika/decision-index-results updated a model 8 days ago
Mapika/decider-chat-qwen3.6-27b published a model 8 days ago
Mapika/decider-chat-qwen3.6-27bOrganizations
models 14
Mapika/decider-12b
Text Classification • 12B • Updated • 279 • 4
Mapika/decider-chat-qwen3.6-27b
Text Classification • 28B • Updated • 47
Mapika/decider-chat-gemma4-31b
Text Classification • 31B • Updated • 464 • 3
Mapika/decider-2b-coherent
Multiple Choice • Updated • 41 • 1
Mapika/decider-2b-GGUF
Text Classification • 2B • Updated • 2.62k
Mapika/decider-4b-GGUF
Text Classification • 4B • Updated • 4.57k • 2
Mapika/decider-2b
Text Classification • 2B • Updated • 325k • 102
Mapika/decider-4b
Text Classification • 4B • Updated • 24.8k • 22
Mapika/decider-35b-a3b
Text Classification • 35B • Updated • 1.58k • 9
Mapika/decider-35b-a3b-nvfp4
Text Classification • 18B • Updated • 887 • 8