Compare models

Pick two to four models. No markup on model usage: you pay the provider's list price.

Side-by-side comparison of Qwen3.8 2.4T-A95B, Sarvam-105B

Attribute	Qwen3.8 2.4T-A95B	Sarvam-105B
Provider	Alibaba Cloud (Qwen)	Sarvam AI
Model id	qwen/qwen3.8-2.4t-a95b	sarvam/sarvam-105b
Status	Available via Safeguard	Available via Safeguard
Input / 1M tokens	$2	Price on request
Output / 1M tokens	$6	Price on request
Price notes	International (Singapore), 0 to 1M input tokens, both modes.	Sarvam lists INR 29.28 input, INR 10.98 cached input and INR 73.20 output per 1M tokens. Prices are in Indian rupees only, so no US dollar price is shown.
Context	Not published	131K
Parameters	Not published	105B
Max output	Not published	Not published
Modality	Text	Text
Weights	Open	Open
Zero data retention	Not stated	Available
Prompts go to	Alibaba Cloud Model Studio (prices shown are for the Singapore / International deployment)	Sarvam AI through its API: the managed environment runs in Microsoft Azure's Central India region and LLM inference on GPU infrastructure in India run by Yotta and NxtGen. A copy of the open weights you host sends nothing to Sarvam.
Retention	Alibaba Cloud states it never uses your data for model training, Model Studio stores data generated from model calls as required by law (no period stated)	Sarvam keeps Model API data until the workspace Owner sets a retention period, and No retention (0 days) is available for Model APIs. Sarvam's terms let it train its own models on inputs and outputs as its Privacy Policy and the law allow, with consent where required, and its Trust Center says one customer's data never trains models for another customer.
Region	Singapore (International) for these prices, other regions such as China (Beijing), Hong Kong, Frankfurt, Virginia and Tokyo have separate price lists	India: Sarvam's managed production environment runs in Microsoft Azure's Central India region.
Price checked	2026-10-10	2026-10-10