Sources
Loading...
Additional media
Loading...
DeepNewz, mobile.
People-sourced. AI-powered. Unbiased News.
01AI_Yi, founded and led by Kaifu Lee, has announced the release of its latest large-scale language model, Yi-1.5, under the Apache 2.0 license on May 13th. The model, which comes in three sizes (34B, 9B, and 6B), has been continuously pre-trained on 500 billion tokens to enhance coding, reasoning, and instruction-following capabilities. The largest version, Yi-1.5-34B, matches or exceeds the performance of Meta's Llama 3 70B in benchmarks. Additionally, 01AI_Yi has introduced the Yi-Large model, available globally via the Yi-Large API and in China through the Yi-Large + Wanzhi productivity product. The company aims to make AI accessible to everyone with these new releases, including the first consumer application, Wanzhi.
On May 13th, @01ai_yi, a large-scale #startup company founded and led by @kaifulee, officially released the closed-source billion-parameter #LargeModel #Yi-Large. https://t.co/LbvC7Pw9kd
On Bloomberg @business earlier today w @DavidInglesTV, @YvonneManTV and @jenzhuscott 30 min before @01AI_Yi team debuted our most powerful Yi-Large LLM to the world. We aim to make AI accessible to everyone by launching the first consumer application "Wanzhi," plus Yi model APIs… https://t.co/DQFbGGS3FY
𝗬-𝗟𝗮𝗿𝗴𝗲 global debut from https://t.co/0cAMgbVI4j today. Experience our 🔥largest model🔥 in multiple ways: Yi-Large API (global) https://t.co/hayxFMrYat Yi-Large API (China) https://t.co/V84aP7IkZa Yi-Large + Wanzhi productivity product (万知 in China)… https://t.co/XUOFATAyv8
