Sarvam Prism decision model
Sarvam Prism returns probabilities for text or JSON questions. Compare its hosted API limits and rupee pricing with the custom FP8 weight package.
By Sarvam AI · Hugging Face package release: 3 Nov 2026
A Sarvam-105B decision fine-tune, offered as custom FP8 weights and a hosted API. It returns option probabilities without generating text. Choice options are sorted by key before scoring. The hosted API and downloadable runtime have different context limits.
Quick answers
What is Sarvam Prism by Sarvam AI?
Sarvam Prism: A Sarvam-105B decision fine-tune, offered as custom FP8 weights and a hosted API.
How much does Sarvam Prism cost?
Sarvam Prism: Published price (vendor-reported): Sarvam API: ₹4 per million input tokens; output free
Who makes Sarvam Prism?
Sarvam Prism: Sarvam AI
Are Sarvam Prism's weights open?
Sarvam Prism: Public weights and prism.py; hosted POST /v2/systemone Licence: Apache-2.0
Facts
- Size
- 105B; about 106 GB of weight files
- Inputs
- Text or JSON; no images
- Context
- Hosted: 32,768 tokens per question; local runtime: 131,072
- Hardware
- Local FP8 runtime needs about 120 GB free GPU memory; tested on B200
- Licence
- Apache-2.0
- Price
- Sarvam API: ₹4 per million input tokens; output free · Compare prices
- Access
- Public weights and prism.py; hosted POST /v2/systemone
Claims and sources
- Sarvam reports 44 ms median and 420 ms p95 on 1,000 serial public-index requests on a B200. vendor reportedsource
- In its 2,000-question shuffle test, fixed-order preprocessing yields zero changed answers; scoring options as sent changes 4.0%. Sorting is part of the tested system. vendor reportedsource
Links
Source: Launch and evaluation → Source: Card, runtime and weights → Source: Hosted API limits and price →
Related projects
No related projects are listed in the registry.