Shaltiel

[Shaltiel @SShmidman]

@maximelabonne Will do! Either way, feel free to DM me and I'll share what we have already :)

[Shaltiel @SShmidman]

@maximelabonne Nice! We're going to release our training datasets in the coming weeks, including human curated preference data - would be happy to share more with you ahead of time

[Shaltiel @SShmidman]

@levelsio Missed opportunity to call the GDPR Cookie Consent Dismisser "Cookie Monster"

[Shaltiel @SShmidman]

@maximelabonne @neural_avb @liquidai @aiDotEngineer @maximelabonne can you DM me? We're pretraining a new model (~105B) and would love to chat / collab 🙏

[Shaltiel @SShmidman]

@maximelabonne @neural_avb @liquidai @aiDotEngineer Would really appreciate a link as well when it's uploaded :)

[Shaltiel @SShmidman]

@Teknium DM'd!

[Shaltiel @SShmidman]

@digitalix The spark boast 1 Petaflops sparse NVFP4, or 512 teraflops dense. Can you find any configuration with any framework where you can reach that training a model?

[Shaltiel @SShmidman]

@ThePrimeagen I personally found Windsurf's autocomplete much more intuitive - agree with everything else :)

[Shaltiel @SShmidman]

@yoavgo We've got our project going on, if you want to hear more :)

[Shaltiel @SShmidman]

@elbeyoglu @elbeyoglu I'm getting some weird results for our website - just getting some metadata with no content. For example, https://t.co/HNkdG62uVn

[Shaltiel @SShmidman]

@elder_plinius @viemccoy Curious to hear about that as well

[NVIDIA AI Developer @NVIDIAAIDev]

How To Adapt AI for Low-Resource Languages with NVIDIA Nemotron https://t.co/mSEF5llUtE

[Shaltiel @SShmidman]

@yuvalmarton @yoavgo @rtsarfaty @yuvalpi @shmidman אפשר לקבל גישה ב-API לאתר שקישרתי

[Shaltiel @SShmidman]

@yuvalmarton @yoavgo @rtsarfaty @yuvalpi @shmidman יש גם מודל פתוח, אבל הוא נוטה יותר להזיות:
https://t.co/wQWgcDQkZK
https://t.co/hLMrTZkDsr

[Shaltiel @SShmidman]

@yuvalmarton @yoavgo @rtsarfaty @yuvalpi @shmidman https://t.co/jMA1AxWW6K

[Shaltiel @SShmidman]

We've come a long way from the initial announcement of the Sovereign LLM project - excited to showcase our results.

We've written up a tutorial sharing how you can train your own Sovereign LLM using Nvidia Nemotron, NeMo Framework, and DGX Cloud Lepton: https://t.co/1Y65PdMOdp

Great work together with @nvidia

[Shaltiel @SShmidman]

Beyond excited to announce the official release of:

🚀 Dicta-LM 3.0: Advancing The Frontier of Hebrew Sovereign LLMs 🔥

Dicta-LM 3.0 is a powerful open-weight collection of LLMs with full Hebrew support.

View the full announcement here: https://t.co/FxPr1j6yBI

🧵

[Shaltiel @SShmidman]

The models were trained on a cluster of over 100 H200 GPUs, on NVIDIA DGX Cloud Lepton. All training was done using the NVIDIA NeMo Framework and the NVIDIA NeMo-RL library. We are extremely grateful to NVIDIA and their technical teams, who made this all possible!

[Shaltiel @SShmidman]

Beyond excited to announce the official release of:

🚀 Dicta-LM 3.0: Advancing The Frontier of Hebrew Sovereign LLMs 🔥

Dicta-LM 3.0 is a powerful open-weight collection of LLMs with full Hebrew support.

View the full announcement here: https://t.co/FxPr1j6yBI

🧵

[Shaltiel @SShmidman]

📄 The technical report is available here: https://t.co/ramBGf1Rh4