Decision models that return calibrated probabilities for typed questions about text or JSON, without generating text.
AI & ML interests
Indian languages, LLM evaluation, benchmarks, multilingual NLP
Recent Activity
View all activity
Organization Card
An independent research lab building small, specialist, open AI models for Indian languages and domains.
models 9
sthanika-ai/Sieve-31B
Text Classification • Updated • 46 • 7
sthanika-ai/Sieve-0.8B
Text Classification • Updated • 56 • 5
sthanika-ai/Sieve-9B-v2
Text Classification • Updated • 96 • 7
sthanika-ai/Sieve-27B
Text Classification • Updated • 119 • 5
sthanika-ai/Sieve-4B
Text Classification • Updated • 147 • 5
sthanika-ai/Sieve-9B
Text Classification • Updated • 102 • 6
sthanika-ai/Sieve-2B
Text Classification • Updated • 5
sthanika-ai/Sieve-2B-v2
Text Classification • Updated • 5
sthanika-ai/gemma3-12b-kcc-advisory
Text Generation • Updated • 18 • 4
datasets 10
sthanika-ai/Sieve-31B-decision-index-results
Updated • 114 • 5
sthanika-ai/Sieve-0.8B-decision-index-results
Updated • 82 • 5
sthanika-ai/Sieve-27B-decision-index-results
Updated • 155 • 4
sthanika-ai/Sieve-9B-v2-decision-index-results
Updated • 175 • 3
sthanika-ai/Sieve-9B-decision-index-results
Updated • 136 • 4
sthanika-ai/Sieve-4B-decision-index-results
Updated • 244 • 5
sthanika-ai/Sieve-2B-v2-decision-index-results
Updated • 117 • 3
sthanika-ai/Sieve-2B-decision-index-results
Updated • 139 • 4
sthanika-ai/Indic-KCC-Agri-Advisory-Benchmark
Viewer • Updated • 5.5k • 99 • 5
sthanika-ai/Bharat-Knowledge-Probe-Benchmark
Viewer • Updated • 552 • 68 • 1