Co-founder https://nitter.gq/t.co/BMcuoqJ1d5, a pioneering AI startup at the forefront of knowledge extraction from diverse sources. LLM and Deep Learning model training.
Athens, Greece
Joined October 2009
- Tweets382
- Following778
- Followers238
- Likes103
V. Antonopoulos retweeted
Say hello to Gemini Embedding 2, our new SOTA multimodal model that lets your bring text, images, video, audio, and docs into the same embedding space! 👀
V. Antonopoulos retweeted
Last week, we presented our work at the "NLP for Space Science" Workshop (European Space Agency, Madrid). Inspiring experience, lots of insightful discussions. Impressed by the motivation within the community, truly grateful to the organizers!
#ESA #SpaceScience #NLP #LLMs
An evaluation of various models as generators in a RAG application handling Greek text. @11tensors new blogpost medium.com/11tensors/are-lla…
#Llama3 #Mixtral #RAG #LLMs
New Mistral model just released!
magnet:?xt=urn:btih:9238b09245d0d8cd915be09927769d5f7584c1c9&dn=mixtral-8x22b&tr=udp%3A%2F%2Fopen.demonii.com%3A1337%2Fannounce&tr=http%3A%2F%https://nitter.gq/t.co/OdtBUsbeV5%3A1337%2Fannounce
V. Antonopoulos retweeted
Introducing Command R Plus ⚡
> Beats claude-3, mistral-large, gpt-4 turbo.
> 104 Billion parameters.
> Built with multi-step tool use and RAG.
> Supports 10 languages.
> Context length of 128K.
> Trained with grounded generation capabilities - citations and responses based on context.
> Optimised for code - rewrites, snippets and explanations.
> Available in 4-bit and 8-bit too.
> Directly available in Transformers to play around. 🤗
huggingface.co/CohereForAI/c…
V. Antonopoulos retweeted
🚀crewAI 0.27.0 is out!!
Super reliable and high quality results
Really appreciate RTs ❤
🧠 Memory long, short and entity
🤚🏼 Native Human Input
🌐 Local models RAG Tools
🔍 Finer Cache Control
⛓️ Improved Cross Agent Collab
📝 Improved tool usage
🔒 Security improvements
More!
LLMs, Embedding models and Deep Learning solutions for downstream Space applications medium.com/11tensors/llms-em…
V. Antonopoulos retweeted
We’re announcing a new optimised model today! Mistral Large has top-tier reasoning capacities, is multi-lingual by design, has native function calling capacities and a 32k model. The pre-trained model has 81.2% accuracy on MMLU.
Learn more on mistral.ai/news/mistral-larg….
Mistral Large is available on la Plateforme, as well as on Azure, following our commitment to bring frontier AI everywhere. We're eager to see what developers do with Mistral Large!
V. Antonopoulos retweeted
An over-enthusiastic employee of one of our early access customers leaked a quantised (and watermarked) version of an old model we trained and distributed quite openly.
To quickly start working with a few selected customers, we retrained this model from Llama 2 the minute we got access to our entire cluster — the pretraining finished on the day of Mistral 7B release.
We've made good progress since — stay tuned!
V. Antonopoulos retweeted
A biLSTM architecture for the creation of semantic links in the scientific publications of the European Space Agency medium.com/11tensors/a-bilst…
V. Antonopoulos retweeted
🐶 NeuralBeagle14-7B
It's the best-performing 7B parameter model on the Open LLM Leaderboard.
Remarkably, it also ranks as the 10th best-performing model overall on the Open LLM Leaderboard. In just 7B parameters!
Merge + DPO = profit
V. Antonopoulos retweeted
[Arena] Exciting update! Mistral Medium has gathered 6000+ votes and is showing remarkable performance, reaching the level of Claude. Congrats @MistralAI!
We have also revamped our leaderboard with more Arena stats (votes, CI). Let us know any thoughts :)
Leaderboard
huggingface.co/spaces/lmsys/…
V. Antonopoulos retweeted
NEW PAPER: Mixtral of Experts 🧡arxiv.org/abs/2401.04088
Check out how Mixtral 8x7B works!
We are also hosting an office hour on @MistralAI Discord tomorrow. Join us if you have any questions about Mistral models or la plateforme: discord.gg/mistralai
V. Antonopoulos retweeted
LLMs for Generative Information Extraction
One of the most common uses of LLMs is to analyze and extract information. It's also probably one of the easiest and most effective ways to use LLMs if done right.
Here is a new survey paper on methods, trends, domains, insights, and tasks where LLMs are effective for information extraction.
arxiv.org/abs/2312.17617v1
V. Antonopoulos retweeted
Fine-tune a Mistral-7b model using Direct Preference Optimization (DPO).
Just published a tutorial on @TDataScience about using DPO to enhance the performance of SFT models. Funnily enough, I created NeuralHermes-2.5 for this article.
towardsdatascience.com/fine-…
V. Antonopoulos retweeted
Mixtral-8x7B on free Google Colab just dropped!
TL;DR:
- Novel offloading trick & mixed quantization.
- Managed to run Mixtral-8x7B on free-tier Google Colab.
Notebook: github.com/dvmazur/mixtral-o…
Code: github.com/dvmazur/mixtral-o…
Paper: arxiv.org/pdf/2312.17238.pdf
V. Antonopoulos retweeted
🚀 Open-source AI strikes again! Announcing Notux 8x7B, a fine-tune of Mixtral Instruct with high-quality chat data and DPO.
Notux now the top ranked MoE on the Open LLM leaderboard.
huggingface.co/argilla/notux…
V. Antonopoulos retweeted
📢Just change the model name, you can run LLMs blazingly fast on your PC using Intel Extension for Transformers powered by SOTA low-bit quantization!
🎯Code: github.com/intel/intel-exten…, supporting Mistral, Llama2, Mixtral-MOE, Phi2, Solar, most recent LLMs.
🤗github.com/intel/intel-exten…
V. Antonopoulos retweeted
LLaMA-Factory
github: github.com/hiyouga/LLaMA-Fac…
Easy-to-use LLM fine-tuning framework (LLaMA, BLOOM, Mistral, Baichuan, Qwen, ChatGLM)