278M Telugu model Vemana trained from scratch for ~$69 beats sub-1B open baselines
pranavmarla · x · 2026-10-27
Vemana is a 278M-parameter Telugu model trained from scratch on a single GPU for about $69, now live.
- On new Telugu Wikipedia pages it beats every open base model under 1B tested, including nanochat-telugu-560M, which is twice its size.
- All weights, every checkpoint, and the eval code are public.
- It runs entirely in the browser on your own device, with nothing sent anywhere.
More from Models
- Founder forgets to turn off 'ultrafast' mode, gets a $57,000 OpenAI bill — joshgonsalves_ · 2026-10-27
- MathArena flags Fireworks inference issues as DeepSeek-V4.1-Flash lands at 49% for $1.29 — teortaxesTex · 2026-10-27
- GPT-6.1 Terra hits 78.5 on Browser Use Benchmark, near Sol at 35% less cost — sandersted · 2026-10-27
- Reddit asks: which 3-8B small model handles real-time EN/CN translation well? — Loose_Comparison368 · 2026-10-27
- Insider says actual V4.1 Pro will be similarly built, current version not materially stronger — teortaxesTex · 2026-10-27
- GPT-6.1 Terra scores best-in-class 8% on MazeBench at half of Sol's cost — patience_cave · 2026-10-27