Mastodawn

InfoQ Feb 2, 2024

Stability AI released two sets of pre-trained model weights for Stable LM 2 - a 1.6B parameter language model.

#StableLM 2 is trained on 2 trillion tokens of text data from seven languages and can be run on common laptop computers.

Read more on #InfoQ: https://bit.ly/3w7IUsC

#AI #ML #LLMs #DeepLearning

Stability AI Releases 1.6 Billion Parameter Language Model Stable LM 2

Stability AI released two sets of pre-trained model weights for Stable LM 2, a 1.6B parameter language model. Stable LM 2 is trained on 2 trillion tokens of text data from seven languages and can be run on common laptop computers.

InfoQ

KINEWS24 Oct 4, 2023

Stability Stable LM 3B now introduces small LLM for mobile devices

#ai #generativeki #languagemodel #nlp #stablelm #opendata #portableki #edgecomputing

https://kinews24.de/stability-stable-lm-3b/

Stability Stable LM 3B stellt jetzt kleines LLM für mobile Geräte vor - KiNews24.de

Stability Stable LM 3B: Ankündigung von Stable LM 3B, einem kompakten 3-Milliarden-Parameter-generativen KI-Modell, das für einen effizienten Einsatz auf tragbaren Geräten mit beeindruckender Leistung in Bezug auf Größe entwickelt wurde.

KI NEWS24

KINEWS24 Oct 3, 2023

Stability Stable LM 3B stellt jetzt kleines LLM für mobile Geräte vor

#ai #ki #generativki #sprachmodell #nlp #stablelm #offenedaten #tragbareki #edgecomputing

https://kinews24.de/stability-stable-lm-3b/

Stability Stable LM 3B stellt jetzt kleines LLM für mobile Geräte vor - KiNews24.de

KI NEWS24

Show thread

Cragsand

Jun 12, 2023

@Ruth_Mottram StabilityLabs who are behind #StableDiffusion have released #StableLM you can run at home. For it to give decent results you have to use the really large checkpoint models which use a lot of VRAM. I'd recommend at least an RTX 3090 with 24gb of VRAM unless you've got an Nvidia CUDA server setup.

An acquaintance has been running it on his PC to generate uncensored fan fiction stories. With #ChatGPT you're limited to their filtered model and API so running it yourself there are no limitations other than requiring a beefy gaming PC.

I haven't tried running it myself since I've got an RTX 2080Ti with 11gb of VRAM. From checking YouTube videos you won't get any good results at all running smaller models.

https://github.com/Stability-AI/StableLM

GitHub - Stability-AI/StableLM: StableLM: Stability AI Language Models

StableLM: Stability AI Language Models. Contribute to Stability-AI/StableLM development by creating an account on GitHub.

GitHub

HhttG May 30, 2023

Spaß mit einer #KI (in dem Fall #StableLM) 😀 :

Historian Matt May 17, 2023

How the Tech Are You episode 46

In this episode:

- Google rolls out passwordless logins with passkeys
- EU allows combustion cars if and only if they use synthetic gas
- StabilityAI releases open source 7B parameter language model for their StableLM AI
- Something Awful community scrambling as Imgur image purge looms
- Discord is moving to more standard handles and some people don’t like it

#Google #StabilityAI #StableLM #Imgur #SomethingAwful #Discord

https://m.youtube.com/watch?v=lqasS5BVK2c&feature=youtu.be

Passkeys, Synthetic Gas, StableLM AI, Something Awful Saves Imgur, Discord Handles - HTTAY 46

YouTube

David Egts May 12, 2023

"Of course, this is just the first version, & right now #GPT4 is way ahead of it. But since it’s #opensource, you can expect #StableLM to evolve at hyperspeed." https://www.fastcompany.com/90884963/stablelm-new-chatgpt

Victor Paléologue May 5, 2023

It did not take long before an open source combination of #Whisper and an #LLM. Here is QuiLLMan: https://github.com/modal-labs/quillman
Here it is a #Vicuna in action, just like #StableLM's.