Ethan B. Holland

Over 56,100 manually organized AI links and counting

a middle school art project showcasing the spirit of sharing, a small written sign says: "Open Source" --chaos 20 --ar 4:3 --style raw --personalize x3p86jc --v 6.1

Open Source AI News: Week Ending 07/19/2024

a middle school art project showcasing the spirit of sharing, a small written sign says: “Open Source” –chaos 20 –ar 4:3 –style raw –personalize x3p86jc –v 6.1

Cohere

Cohere and Fujitsu Announce Strategic Partnership To Provide Japanese Enterprise AI Services

https://cohere.com/blog/fujitsu-partnership

DeepSeek

“🎉Exciting news! We open-sourced DeepSeek-V2-0628 checkpoint, the No.1 open-source model on the LMSYS Chatbot Arena Leaderboard @lmsysorg. Detailed Arena Ranking: Overall No.11, Hard Prompts No.3, Coding No.3, Longer Query No.4, Math No.7. DeepSeek-V2-0628 is released at 

[AINews] DeepSeek-V2 beats Mixtral 8x22B with >160 experts at HALF the cost • Buttondown

https://buttondown.com/ainews/archive/ainews-deepseek-v2-beats-mixtral-8x22b

Meta/Llama

“From this: – Llama 4 started training in June – Llama 4 will be fully multimodal, including audio – Llama 3 405b will still release in the EU – Llama 4, and beyond, will not be released in the EU unless they allow META to use EU Facebook and Instagram posts as training data. 

Introducing Llama-3-Groq-Tool-Use Models – Groq is Fast AI Inference

We are excited to announce the release of two new open-source models specifically designed for tool use: Llama-3-Groq-70B-Tool-Use and Llama-3-Groq-8B-Tool-Use, built with Meta Llama-3. 

https://wow.groq.com/introducing-llama-3-groq-tool-use-models

“Groq’s open-source & synthetic-only model beats closed models! 🚀🧬 Groq’s Llama, trained solely on synthetic data, tops the function calling leaderboard: • 70B version: 90.76% accuracy • 8B version: 89.06% accuracy • Outperforms GPT-4 and Claude.  “By achieving top performance using only synthetic data, Groq challenges the notion that vast amounts of real-world data are necessary for creating cutting-edge AI models.”

Groq’s open-source Llama AI model tops leaderboard, outperforming GPT-4o and Claude in function calling | VentureBeat – https://venturebeat.com/ai/groq-open-source-llama-ai-model-tops-leaderboard-outperforming-gpt-4o-and-claude-in-function-calling/ 

Mistral

Mistral NeMo | Mistral AI | Frontier AI in your hands

https://mistral.ai/news/mistral-nemo

Mistral AI and NVIDIA Unveil Mistral NeMo 12B, a Cutting-Edge Enterprise AI Model | NVIDIA Blog

https://blogs.nvidia.com/blog/mistral-nvidia-ai-model

“Mistral NeMo is out! 🔥😱 A new exciting model from Mistral 👐NVIDIA – 12B model – 128k context length – Apache 2.0 license – Base + instruct-tuned – Multilingual – Function calling Exciting release 🥳 Announcement: 

“Mistral NeMo looks to be a damn good model – 12B so it fine-tunes quickly and cheaply – fast inference (small + trained w/ quantisation awareness) – new tokenizer that handles many languages efficiently 

“Very happy to release our new small model, Mistral NeMo, a 12B model trained in collaboration with @nvidia. Mistral NeMo supports a context window of 128k tokens, comes with a FP8 aligned checkpoint, and performs extremely well on all benchmarks. Check it out! 

“Mistral has released NeMo, a new open-source, long context, smaller model as a successor to Mistral 7B. Why it is exciting below 👇 – Open source model with a 128k context window: Large context window open-source models have been few and far between and the most popular models, 

“Mistral releases 12B open LLM! 🤯 Mistral Nemo comes as a base and instruct version with a 128k Context window and is multilingual in 9 Language with a new tokenizer. 👀 TL;DR: 🧠 12B Base and Instruct drop-in-replacement for Mistral 7B 🪟 Supports 128k context window and new 

“Today we are releasing two small models: Mathstral 7B and Codestral Mamba 7B. On the MATH benchmark, Mathstral 7B obtains 56.6% pass@1, outperforming Minerva 540B by more than 20%. Mathstral scores 68.4% on MATH with majority voting@64, and 74.6% using a reward model. Codestral 

“Somehow this got passed over in my feed today; Mistral released a new base model, slams l3 8b for 4B params more – not sure if its the same arch as old mistrals as its called Mistral Nemo: 

Codestral Mamba | Mistral AI | Frontier AI in your hands

https://mistral.ai/news/codestral-mamba

MathΣtral | Mistral AI | Frontier AI in your hands

https://mistral.ai/news/mathstral

Other Open Source News

“Code released for Robot Synesthesia: In-Hand Manipulation with Visuotactile Sensing! Project page: 

“Researchers open-sourced the code for ‘Robot Synthesia’ It’s an approach to achieving in-hand object rotation using visual/touch inputs The policy is trained in simulation and then transferred to the robot hand without real-world training” 

“Introducing SmolLM: a new SOTA series of 135M, 360M and 1.7B models, perfect for on-device deployment! 🔥 We also uploaded ONNX weights for the models, meaning they can run locally in your browser with 🤗 Transformers.js and WebGPU acceleration! ⚡️ Try it out (+ blog post)! 👇 

“On-device deployment of LLMs is more important than ever. Today we’re releasing SmolLM a new SOTA series of 135M, 360M and 1.7B models: – Outperforming MobileLLM, Phi1.5 and Qwen2 small models – Trained on SmolLM-corpus, of high quality web, code and synthetic data 

SmolLM – blazingly fast and remarkably powerful

https://huggingface.co/blog/smollm

“Today we are announcing a new inference stack, which provides decoding throughput 4x faster than open-source vLLM. We are also introducing new Together Turbo and Together Lite endpoints that enable performance, quality, and price flexibility so you do not have to compromise. 

“Open-source models winning Math Olympiads and solving viral math problems on Twitter 

“Patronus AI announced the release of ‘Lynx’. It’s a new open-source hallucination detection model. The startup claims it outperforms existing solutions such as GPT-4, Claude-3-Sonnet, and other models used as judges in closed and open-source settings 

Introducing the Coalition for Secure AI, an OASIS Open Project – OASIS Open

Introducing the Coalition for Secure AI, an OASIS Open Project

“• Uncensored Models: Mythomax-l2-13b and Chronos-hermes-13b generated detailed responses without filtering, demonstrating fewer content restrictions. https://twitter.com/premai_io/status/1813217404606161131

Heads up! You’ve scrolled to the end of this category. There may have been just one or two links (above), so go back up and double check to be sure you didn’t quickly scroll down past it.

Be Sure To Read This Week’s Main Post:

This week’s executive overview and top links are here:

AI News #42: Week Ending 07/19/2024 with Executive Summary and Top 58 Links 

The post you just read is an deep dive extension of my weekly newsletter, This Week In AI, an executive summary of the top things to know in AI. Each week, I create an accessible overview for laypeople to feel confident they are conversant with the week’s AI developments. I include a curated list of must-click links of the week, to offer everyone a hands-on opportunity to explore the most intriguing updates in artificial intelligence across various categories, including robotics, imagery, video, AR/VR, science, ethics, and more. Beyond the overview, I post these topic-based deeper dives (below). If you haven’t read this week’s overview, I recommend starting there.

Credits/Sources

Most of these weekly links come from just a few prolific oversharing sources. Please follow them, as they work hard to find the news each week and they make it a lot easier for me to compile.

For previous issues, please visit the archives!

Thanks for reading!

Leave a Reply

Trending

Discover more from Ethan B. Holland

Subscribe now to keep reading and get access to the full archive.

Continue reading