
Gemma 4: Now up to 3x Faster. ⚡ Same quality, way more speed. Our new MTP drafters allow Gemma 4 to predict multiple tokens at once, effectively tripling your output speed without compromising intelligence.
ryul99 율 🐲
614 posts

@ryul99_dev
Interested in ML(Deep Learning), Linux ⭐ my tweets are my own

Gemma 4: Now up to 3x Faster. ⚡ Same quality, way more speed. Our new MTP drafters allow Gemma 4 to predict multiple tokens at once, effectively tripling your output speed without compromising intelligence.

I've started a company: ggml.ai From a fun side project just a few months ago, ggml has now become a useful library and framework for machine learning with a great open-source community


영어로 프롬프트를 적으면 성능이 더 나을 것이라 생각해서 + 익숙해지기 위해서 영어로 프롬프트를 작성했었는데요, 뉘앙스 차이가 전달이 안되더라구요. 그래서 prompt rewriter를 시도하던 중 jiun.dev/posts/claude-h… 을 보고 영어학습용으로 방향을 바꾸어보았습니다. torch.vision/posts/claude-e…







🐉Six Dragons Fly Again: It's time for applying symolic music gen to Korean traditional music🇰🇷 Our work on generating Korean court music will be presented @ISMIRConf🥳 Work by @DanbiNaerinHan @MarkGothamMusic DongMin Kim, @crescent_stdio @is_s_yun Web Demo and Concert video⬇️


Six Dragons Fly Again for the Web: An Interactive Web Application for Everyone @ISMIRConf I developed an interactive web demo for the Six Dragons project that can create an ensemble of Korean court music using our project's model.🐉🎵 Implementation: crescent.studio/project/six-dr…








Statement from Scarlett Johansson on the OpenAI situation. Wow: