Cointime update: Indian AI firm Sarvam opened its Token Factory inference platform to developers and enterprises this week, hosting open-weight models including DeepSeek, Moonshot’s Kimi, Z.ai’s GLM, Alibaba’s Qwen and Sarvam’s own models on GPUs located in India, priced in rupees. Sarvam said DeepSeek V4.1 Flash processes over 450 billion tokens a day, with input priced from 14 rupees per million tokens. Co-founder Pratyush Kumar said the platform runs on company-purchased GPUs and that subsidized compute ended in March 2026.
Sarvam Opens Token Factory Inference Platform on India GPUs
Share on WeChat
Open WeChat → Scan → then tap "…" to send to a chat or Moments.
Tap "…" in the top-right corner to send to a chat or share to Moments.
Related News
8m ago
China chip industry sales forecast at 1.58 trillion yuan in 2026
10m ago
IMF’s Georgieva: AI investment underpins global economy, APEC growth seen at 3.4%
17m ago
Hyperscale Data says Michigan AI project needs $74.8M more
34m ago
Lenovo Q2 revenue tops RMB200B for first time, adjusted net profit up 166%
57m ago
OpenAI phone project taps Foxconn unit FIH for assembly
1h ago
Nvidia’s $500B AI Financing Plan Has Deployed No Capital Yet