Latest
  1. Johns Hopkins Bayview tells 3,182 patients an employee put their health data into a public AI tool
  2. China’s AI compute access expected to double in 2027, with US-designed chips about 65 per cent
  3. UK commits £100m to national AI Biosecurity Lab
  4. Moonshot AI releases Kimi K3.1 Pro with 1-million-token context window
  5. Residents in Newport and Cardiff oppose hyperscale datacentres near homes
  6. Study: Grok 4.7 picked Republicans in 92% of forced midterm answers, GPT-6.1 Sol picked Democrats in 97%
  7. Attackers exploiting the PHP flaws hand the break-in to AI agents, Unit 42 says
  8. Elvis Presley’s rightsholders sue Anthropic over training on 274 songs

Latest / Frontier models and safety

Model releaseFrontierIndiaConfirmed

Sarvam releases 30B and 105B open models trained from scratch in India

See how this story connects

Source: Sarvam AI documentationAlso reported by Business Today

Sarvam AI launched Sarvam-30B and Sarvam-105B at the India AI Impact Summit in New Delhi on 18 February 2026. Sarvam-105B is a mixture-of-experts model with 128 experts and a 128,000-token context, released under the Apache 2.0 licence and covering English and 10 Indian languages; Sarvam-30B was pre-trained on 16 trillion tokens. Sarvam says both were trained from scratch.

Why it matters

Developers in India get openly licensed models built for Indian languages from the lab chosen under the IndiaAI Mission, usable without dependence on foreign model providers.

When
18 Feb 2026
ID
WTA-N-2026-0107
Where
India
Who
Sarvam AI, IndiaAI; Pratyush Kumar
Rules involved

Summary by wheresthe.ai, from Sarvam AI documentation and the reporting credited above. Checked against the source on 9 Oct 2026.

Read the original at Sarvam AI documentation

Source copy: download the page as saved on 18 Oct 2026 (PDF). Also held by the Internet Archive.

Next stop on this Storyline: What happened nextNew Delhi Declaration on AI Impact endorsed by 88 states and bodies, including the US and China

Cite this story

WTA-N-2026-0107: “Sarvam releases 30B and 105B open models trained from scratch in India.” wheresthe.ai, 18 Feb 2026. https://wheresthe.ai/d/in-sarvam-30b-105b-open-models-2026-02/ (source: Sarvam AI documentation).

Spotted an error? Email [email protected].