Pierre Stock (@PierreStock) / X

Pierre Stock

165 posts

Pierre Stock

@PierreStock

VP of Science Operations @mistralai. Ex RS @metaAI | PhD @ENSdeLyon | MVA @ENS_ParisSaclay

Paris, France

scholar.google.fr/citations?user…

Joined July 2010

Pinned
Pierre Stock
@PierreStock
May 29, 2024
Super excited to release Codestral 22B, our first-ever code model @MistralAI, capable of understanding 80+ programming languages. Codestral is way smaller and faster than its competitors while performing similarly or even better! Blogpost: lnkd.in/eGK66TMN
14K
Pierre Stock
@PierreStock
Jun 15, 2021
Excited to (re)-join Facebook AI as a Research Scientist after successfully defending my PhD around Efficiency and Redundancy in Deep Learning Models!
Pierre Stock
@PierreStock
Jul 28, 2019
Our latest paper on network compression is available! We manage to get a 76.1% top-1 ResNet-50 that fits in 5 MB and also compress a Mask R-CNN within 6 MB. Code and compressed models: github.com/facebookresear… Paper: arxiv.org/abs/1907.05686
AI at Meta
@AIatMeta
Jul 26, 2019
Facebook AI researchers have developed a new method for reducing the memory footprint of neural networks by quantizing their weights, while maintaining a short inference time. #AI ai.facebook.com/blog/compressi…
GitHub - facebookresearch/kill-the-bits: Code for: "And the bit goes down: Revisiting the quantiz...
From github.com
Pierre Stock
@PierreStock
Feb 26, 2024
Extremely excited by our new release @MistralAI: - Mistral Large: our new flagship model: mistral.ai/news/mistral-l… - Le Chat: a chat interface for our model family chat.mistral.ai - Function calling, JSON mode, new very efficient small endpoint and way more!
34K
Pierre Stock
@PierreStock
Dec 11, 2023
Mixtral 8x7B is here, 11 weeks only after Mistral 7B. Outperforms Llama 2 70B and GPT 3.5 on most benchmarks, at the inference cost of a 12B dense model, with 32k tokens context size.
48K
Pierre Stock
@PierreStock
Sep 27, 2024
First multimodal model, a 12B, on par with the other big ones. And we’re not done cooking!
Arena.ai
@arena
Sep 27, 2024
Exciting update from Vision Chatbot Arena! We’ve gathered over 6K new votes for the latest open vision models (Qwen2, Llama 3.2, Pixtral) and ChatGPT-4o. - ChatGPT-4o has taken the #1 spot, surpassing Gemini. - Open models (Qwen, Llama 3.2, Pixtral) are rapidly improving,
26K
Pierre Stock
@PierreStock
Apr 23, 2020
Neural models have great performance, but how can we use them in practice? Excited to share our latest work on quantization, where we develop a technique to aggressively compress vision and NLP models while maintaining strong performance. Code released too!
AI at Meta
@AIatMeta
Apr 23, 2020
We are releasing code and sharing details for Quant-Noise, a new technique to enable extreme compression of state-of-the-art #NLP and computer vision models without significantly affecting performance. ai.facebook.com/blog/training-…
00:00
Pierre Stock
@PierreStock
Jul 2, 2024
Thrilled to see start-ups like Pretzel AI build products on top of our code model @MistralAI, dubbed Codestral!
pretzelai/README.md at main · pretzelai/pretzelai
From github.com
7.8K
Pierre Stock
@PierreStock
Sep 27, 2023
Super proud to release our very first model, the best 7B in town, after an intense summer at @MistralAI!
Guillaume Lample @ NeurIPS 2024
@GuillaumeLample
Sep 27, 2023
Mistral 7B is out. It outperforms Llama 2 13B on every benchmark we tried. It is also superior to LLaMA 1 34B in code, math, and reasoning, and is released under the Apache 2.0 licence. mistral.ai/news/announcin…
4.6K
Pierre Stock
@PierreStock
Mar 1, 2019
Check out the final version of our ICLR 2019 paper "Equi-normalization of Neural Networks"! Paper: arxiv.org/abs/1902.10416 Code: github.com/facebookresear…
Pierre Stock
@PierreStock
Jun 15, 2021
Replying to @PierreStock
PhD manuscript available here: theses.fr/2021LYSEN008
Pierre Stock
@PierreStock
Oct 16, 2024
Ministral 3B and 8B, your perfect models for edge/agentic use-cases!
Guillaume Lample @ NeurIPS 2024
@GuillaumeLample
Oct 16, 2024
We just released two small models, with 3B and 8B parameters. Ministral 3B is exceptionally strong, outperforming Llama 3 8B and our previous Mistral 7B on instruction following benchmarks. mistral.ai/news/ministrau…
2.8K
Pierre Stock
@PierreStock
Jun 5, 2024
My Tailor is Mistral! Model Customization API: - Finetune a Mistral model efficiently with LoRA in a few clicks - Get a cost-effective personnalized API endpoint It's been (exactly) a year since we started and it's only the beginning! Blog post: lnkd.in/gFQnUERB
3.5K
Pierre Stock
@PierreStock
Sep 17, 2024
Pixtral is out, try it out to analyze images as input!
Mistral AI
@MistralAI
Sep 17, 2024
Replying to @MistralAI
mistral.ai/news/pixtral-1… 2/2
1.3K