udgamlabs/Vemana-Base

just landed · new, unvettedmodel

New, unvetted: created or first seen in the last few days, not from a trusted org and not past the signal threshold yet.

Vemana is a 278M-parameter Telugu language model, trained from scratch on 5.1 billion tokens of public Telugu web text, read about 2.4 times (12.0 billion token positions), with its own 32,000-token Telugu tokenizer. Dense, 278M params. Context: 4K tokens (config). License: apache-2.0.

Org

udgamlabs

Released

2026-10-27

Status

just landed · new, unvetted

License

apache-2.0 (permissive)

Parameters

278M

Architecture

dense · Gemma4ForCausalLM

Context

4,096 tokens

Task

text-generation · text

Signal

6 HF likes · 2 downloads

Runs on

real quant files; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.

• Q4 ~1.3 GB: no listed device
• Q8 ~1.4 GB: no listed device
• FP16 ~1.7 GB: 8GB GPU (4060 / 3070), 12GB GPU (RTX 3060), 16GB GPU (5060 Ti / 4060 Ti / 4080 / 5080), 24GB GPU (3090 / 4090), 32GB GPU (5090), 48GB GPU (RTX 6000 Ada / A6000), 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), 16GB GPU (7800 XT / 9070 XT), 24GB GPU (7900 XTX), Strix Halo 128GB unified (~96GB usable), Arc B60 24GB, Arc B70 32GB, Mac 16GB, Mac 24GB, Mac 36GB, Mac 48GB, Mac 64GB (M1 Max), Mac 96GB, Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 32GB RAM (slow), CPU only, 64GB RAM (slow), CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)

Links

• Hugging Face
• GGUF: udgamlabs/Vemana-Base-GGUF

First published here 2026-10-27 · updated 2026-10-27 · markdown · JSON