Chinese lab DeepSeek has released preview details for DeepSeek V4, a major update the company says brings its models much closer to so-called frontier systems. The announcement introduces two variants — V4 Flash and V4 Pro — and highlights big increases in scale, a 1 million-token context window, and aggressive pricing that positions the models as lower-cost alternatives to high-end
Tag: AI benchmarks
Sarvam AI Emerges as a Global Contender, Outperforming Gemini and ChatGPT
India’s Sarvam AI has rapidly gained global attention by delivering top-tier performance on key India-specific AI tasks, demonstrating that focused, localized engineering can outmatch broader, general-purpose models. Sarvam Vision: OCR breakthroughs Sarvam Vision, the company’s optical character recognition model, has achieved standout results on established benchmarks. It posted an accuracy score on olmOCR-Bench that surpasses several high-profile models, and it