---
title: sarvamai/prism
url: https://recentmodel.com/models/sarvamai/prism/
status: just landed · new, unvetted
published: 2026-11-02T11:28:35Z
updated: 2026-11-04T00:22:00Z
source: https://huggingface.co/sarvamai/prism
---

# sarvamai/prism

Sarvam-Prism is a decision model built on Sarvam-105B. Mixture-of-experts, 105B total (128 experts, 8 per token). Context: 128K tokens (config). License: apache-2.0.


- **Org:** sarvamai
- **Released:** 2026-11-02
- **Status:** just landed · new, unvetted
- **License:** apache-2.0 (permissive)
- **Parameters:** 105.0B
- **Architecture:** MoE · 128 experts, 8/token · SarvamMLAForCausalLM
- **Context:** 131,072 tokens
- **Task:** text-classification · text
- **Signal:** 18 HF likes · 0 downloads

## Runs on

estimate from params; sizing calculation, not tested.

- Q4 ~64.3 GB (estimated): 96GB GPU (RTX Pro 6000 Blackwell), Strix Halo 128GB unified (~96GB usable), Mac 96GB, Mac 128GB
- Q8 ~112.9 GB (estimated): 1× DGX Spark (128GB), Mac 192GB, Mac 256GB, CPU only, 128GB RAM (slow)
- FP16 ~211.3 GB (estimated): 2× DGX Spark (256GB), Mac 512GB, CPU only, 256GB RAM (slow)

## Links

- [Hugging Face](https://huggingface.co/sarvamai/prism)