navya-0
Navya is an India-first financial foundation model family, trained from scratch on an India-first corpus (finance + English + Hindi/Hinglish and other Indian languages), custom 64k tokenizer.
| Parameters | ~57.7M |
| Training tokens | not disclosed |
| Trained | 2026 |
| Type | base |
| Author | Navyam AI (Bachatt) |
| License | Apache-2.0 |
| Code | https://github.com/bachatt-app/navyam-gpt |
The earliest from-scratch Navya base model — a proof-of-recipe run.
Intended use: India personal-finance Q&A. Research model — not investment advice.
Usage
from transformers import AutoModelForCausalLM, AutoTokenizer
tok = AutoTokenizer.from_pretrained("navyam-ai/navya-0")
model = AutoModelForCausalLM.from_pretrained("navyam-ai/navya-0")