How Sarvam launched Sarvam Token Factory on X
Serves open-weight AI models from India-hosted infrastructure for enterprises and developers
What Sarvam launched

Today, we are launching Sarvam Token Factory for everyone. We host our own models, and the full range of best-in-class open-weight models, in India. We match leading inference providers on performance and reliability. For enterprises, this means more models to choose from, https://t.co/RMqEQIFqdC
How LaunchVideo.com rated the launch
LaunchVideo.com rating
About the launch
Sarvam has opened Token Factory, an inference service that hosts the company's own models and a set of open-weight models on GPU clusters in India. Enterprises and developers call them through one OpenAI-compatible API. Switching models is a one-line change to the model field, and the service uses the same API key, SDK, and credits as Sarvam's other APIs. Co-founder Pratyush Kumar has said that data residency and a lack of in-country inference have kept many Indian teams off strong open models. On Token Factory, prompts run on Sarvam-operated GPUs in India and are not forwarded to the original model developer's overseas API. The announced catalogue includes Sarvam 105B, GLM 5.3, GLM 5.3 Flash, DeepSeek V4.1 Flash 1017, DeepSeek V4 Pro, Kimi K3, Kimi K3.1 Pro, and Gemma 4 31B. Sarvam 105B Conversations, listed for real-time dialogue and voice agents, carries the same introductory rate as Sarvam 105B: ₹10 per million input tokens and ₹25 per million output tokens through the end of 2026. DeepSeek V4.1 Flash is listed at ₹14 per million input tokens and ₹55 per million output tokens over the same period. Sarvam says that on most of the models it serves, speed and cost match or beat leading global inference providers, based on the company's own measurements from Mumbai. Zero data retention is available on request, and Token Factory carries a 99.95% uptime SLA. The same inference stack can also be run on premises. Sarvam, based in Bengaluru, was founded in August 2023 by Dr. Vivek Raghavan and Dr. Pratyush Kumar. Raghavan has worked on India's digital public infrastructure, and Kumar has led open-source AI work for Indian languages.
Tags
Questions
Frequently asked questions
The Sarvam launch post reached 373K views on X, with 2.2K likes, 269 reposts, 117 comments, 397 bookmarks.
Sarvam launched on X on November 16, 2026 with the message "Serves open-weight AI models from India-hosted infrastructure for enterprises and developers".
Sarvam has opened Token Factory, an inference service that hosts the company's own models and a set of open-weight models on GPU clusters in India.
LaunchVideo.com rated the Sarvam launch 3.80 out of 5: hook 3/5, editing / creativity 4/5, copy 4/5, sentiment of launch 5/5, distribution strategy 3/5.







