Sarvam GGUF models
111 GGUF models in the Sarvam family, quantized by 56 uploaders. Sizes run from 13M to 106B parameters. The longest trained context is 256K tokens.
About the Sarvam GGUF models
Sarvam is represented on ggufmodels.com by 111 GGUF models, quantized by 56 uploaders on Hugging Face. The most downloaded are sarvam 30b GGUF IQ2 M IQ1 M, sarvam 105b GGUF IQ2 M IQ1 M and Sarvamai.Sarvam 30b. Parameter counts run from 13M to 106B, and the longest trained context window is 256K tokens.
A Q4_K_M file of a Sarvam model is between 1.5 to 64.2 GB, which is the quant most people start with. Most files come from tifin-india (12), Savyasaachin (11) and mradermacher (8). The quants you will meet most often are Q4_K_M (46), Q8_0 (39), Q6_K (26) and Q5_K_M (22), where the number is how many models offer that quant.
Frequently asked questions
Which Sarvam GGUF should I download?
The most downloaded Sarvam GGUF here is sarvam 30b GGUF IQ2 M IQ1 M. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.
How much VRAM does a Sarvam model need?
Q4_K_M files of Sarvam models on this site range from 1.5 to 64.2 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.
Where do Sarvam GGUF files come from?
The most common quantizers here are tifin-india, Savyasaachin and mradermacher. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.