Let me introduce you to a company, which didn’t exist just 8 months ago, and has managed to shake the entire AI industry in that short period.

Also, to read:

https://medium.com/@pareto_investor/apple-unveils-ferret-its-first-open-source-ai-surpassing-gpt-4-66ec27642dfc?source=post_page-----bcee41843775--------------------------------

They’ve released alternative models to ChatGPT that have surpassed all competition, valuing the company at nearly $2 billion.

Remarkably, they’ve achieved this without any deceptive promotional videos or major marketing campaigns.

Their work is so intriguing that I find myself checking Twitter daily, just to see if they’ve made any new announcements!

Let me explain why these French innovators have completely transformed the game and how you could also benefit from their advancements.

To start, I suggest we look at a table of the best AI models competing with ChatGPT.

LMSys Leaderboard. (Screenshot from Dec 22, 2023) Mixtral 8x7B Instruct v0.1 achieves an ArenaElo rating of 1121 outperforming Claude-2.1 (1117), all versions of GPT-3.5-Turbo (1117 best), Gemini Pro(1111), and Llama-2–70b-chat (1077). Mixtral is currently the best open-weights model by a large margin.

This table reveals some fascinating points.

For instance, it seems ChatGPT shows regression between versions.

There are also scores that appear inconsistent or out of order, which is odd.

But most notably, there are these small yellow lines labeled with names like “Mixtral”— names that evoke the wind.

At first glance, it doesn’t seem very impressive, ranked seemingly low.

Let me introduce you to a company, which didn’t exist just 8 months ago, and has managed to shake the entire AI industry in that short period.

Also, to read:

https://medium.com/@pareto_investor/apple-unveils-ferret-its-first-open-source-ai-surpassing-gpt-4-66ec27642dfc?source=post_page-----bcee41843775--------------------------------

They’ve released alternative models to ChatGPT that have surpassed all competition, valuing the company at nearly $2 billion.

Remarkably, they’ve achieved this without any deceptive promotional videos or major marketing campaigns.

Their work is so intriguing that I find myself checking Twitter daily, just to see if they’ve made any new announcements!

Let me explain why these French innovators have completely transformed the game and how you could also benefit from their advancements.

To start, I suggest we look at a table of the best AI models competing with ChatGPT.

LMSys Leaderboard. (Screenshot from Dec 22, 2023) Mixtral 8x7B Instruct v0.1 achieves an ArenaElo rating of 1121 outperforming Claude-2.1 (1117), all versions of GPT-3.5-Turbo (1117 best), Gemini Pro(1111), and Llama-2–70b-chat (1077). Mixtral is currently the best open-weights model by a large margin.

This table reveals some fascinating points.

For instance, it seems ChatGPT shows regression between versions.

There are also scores that appear inconsistent or out of order, which is odd.

But most notably, there are these small yellow lines labeled with names like “Mixtral”— names that evoke the wind.

At first glance, it doesn’t seem very impressive, ranked seemingly low.