Nonprofit lab releases Agaram 2.5B, the first Tamil model trained from scratch with tool calling
SuB8u · x · 2026-10-29
MuraiLabsHQ, a small nonprofit lab, has released Agaram 2.5B, a Tamil-first language model trained from scratch — reportedly the first from-scratch Tamil model capable of tool calling. It chats in Tamil, Tanglish and English and handles translation.
The base model, chat model and GGUF files are all open-sourced under Apache-2.0 and can run offline on a laptop. After a four-day public preview of the first chat model, the team rebuilt its fine-tuning around real-user feedback using 222,662 records.
More from Models
- Google unveils Gemini 4 Carbon, tops Artificial Analysis index at 59 two weeks after Argon — jyangballin · 2026-10-29
- Gemini 4 Carbon benchmarks prompt TheZvi to declare Gemini 4 Argon 'was' near-frontier — TheZvi · 2026-10-29
- Preprint: Human peer reviewers caught only 27% of major errors, all 15 AI models beat them (53-96%) — anshulkundaje · 2026-10-29
- Study: top reasoning effort costs up to 42x more on tasks low effort already solves — zainhas · 2026-10-29
- Giotto launches as 'European sovereign AI,' gets exposed as Kimi-based with benchmark EULA gag — TheZachMueller · 2026-10-29
- Google Announces Gemini 4 Carbon, Its Next Frontier Model, Starting With Cyber Defenders — Snoo26837 · 2026-10-29