EKA-1
Soket - Nov 3, 2026
26.3B-parameter open sparse hybrid foundation model trained on ~7.5T tokens across all 22 scheduled Indian languages plus English, for math, code, reasoning and agentic use (IndiaAI Mission). Early access via waitlist.
- Access
- Open source
- Context
- —
- Parameters
- 26.3B
- Source
- open