Shisa.AI

Hi everyone!

With so many languages worldwide, fine-tuning models to achieve SOTA performance in specific languages is becoming incredibly important for truly global AI. Shisa.AI's latest work with their Shisa V2 405B model is a powerful example of this, especially for Japanese.

Shisa.AI has developed this new open-source model, built on Llama 3.1 405B, and it's delivering impressive results. They report it not only surpasses previous GPT-4 versions in their Japanese/English evaluations but also competes head-to-head with the latest models like GPT-4o and DeepSeek-V3 on Japanese benchmarks. A key to this success, they emphasize, was high-quality data.

Beyond the massive 405B model itself, Shisa.AI has also open-sourced their core Shisa V2 JA/EN synthetic dataset, which they believe can boost Japanese capabilities in almost any base model. You can download the model and dataset, and even chat with an FP8 version now.

Shisa V2 405B 日本語上手！

🌏 Open-source Japanese LLM foundation? This could be huge for:

- Localized AI applications 🗼

- Low-resource language support 💎

- Cultural nuance preservation 🎎

The tokenizer design for agglutinative grammar will be critical.

Huge congratulations on the release! Shisa.AI is a major step forward for open-source bilingual models — and it’s great to see such strong performance on Japanese benchmarks. Excited to see how the community builds with Shisa V2 405B!

Shisa.AI’s Shisa V2 405B is a powerful tool for anyone working with bilingual Japanese/English language models! Built on Llama 3.1 405B, it rivals GPT-4o and DeepSeek-V3 on Japanese tasks, offering advanced capabilities for language processing. I’m excited to see how its release, including the model, dataset, and chat demo, can elevate AI-driven Japanese language tasks!

soooooooooo CUTE! Congrats on that big launch guys! 🥰👏