What if running modern models didn’t mean juggling tools? llamafile 0.10.0 is a full rebuild. You get portability and bundled models, plus multimodal, tool calling, and Anthropic Messages API support, all in one executable. See what changed: link.mozilla.ai/llamafile-v010
Davide Eynard
2.4K posts

Davide Eynard
@aittalam
Dad of https://t.co/PvsS2MPEJI, PicoGopher, and two amazing kids. Genetically a teacher, forever a student. He/him

What if running modern models didn’t mean juggling tools? llamafile 0.10.0 is a full rebuild. You get portability and bundled models, plus multimodal, tool calling, and Anthropic Messages API support, all in one executable. See what changed: link.mozilla.ai/llamafile-v010




Taming randomness in ML models with hypothesis testing and @marimo_io blog.mozilla.ai/taming-randomn…







We are hosting on 21-22 Aug 'Math of GDL' workshop in ICIAM 2023 (largest applied math congress, held every 4 yrs). Welcome to join the meeting in person or via zoom! mathgdl.github.io iciam2023.org Keynotes: @epomqo @guidomontufar @crisbodnar @Francesco_dgv



Transformers are fast at training but slow at inference due to sequential autoregressive decoding, right? Well, not actually. 💥In our recent paper, we introduce Parallel Decoding as a way to decode multiple tokens and speed up existing transformer models for MT up to 2x!

🚀Today, I successfully defended my PhD thesis on Topological Deep Learning @Cambridge_CL📖. I was honoured to discuss my work with my examiners, @wellingmax and @jmhernandez233. Very grateful to everyone who has been part of this journey and in particular to @pl219_Cambridge!




Ritwitta se l'hai visto






