To identify the most underrated AI lab of 2024, several candidates stand out based on their innovative contributions and relatively low media visibility. Here are some notable mentions:

• Mistral: Known for its Mistral Large 2 model, which is significantly smaller than Meta's Llama 3.1 yet offers comparable performance. Despite its capabilities, Mistral has received minimal press coverage, making it a strong contender for being underrated.

• DeepSeek AI: This lab focuses on advanced AI research but lacks the media attention that larger players receive. Its innovative approaches in AI development have not been widely recognized, which positions it as an underrated entity in the field.

• Sakana AI: Founded by a co-inventor of the transformer architecture, Sakana AI is working on new AI architectures that aim to improve efficiency and scalability. Despite its promising innovations, it has not garnered significant media attention, which could classify it as underrated.

• Liquid Neural Networks (LNNs): Developed at MIT, LNNs offer a novel approach to AI that adapts dynamically to data, enhancing efficiency. This innovative technology has the potential to reshape AI applications but remains under the radar in terms of public recognition.

These labs exemplify strong research and innovation while lacking the widespread recognition enjoyed by larger organizations like OpenAI or Google DeepMind.

These odds add up to way more than 100%, y'all can do better than this lol

I think either Alibaba or Qwen should be removed, as Qwen is the name of the model family developed by Alibaba.

Deepseek definitely fulfills this criteria, top 3 coding model overall, all models open source + the cheapest API on the market (cheaper than gpt 4o mini)

https://huggingface.co/spaces/bigcode/bigcodebench-leaderboard

https://aider.chat/docs/leaderboards/

Market resolution:

Market will resolve based on your votes.

What does that mean, exactly?

It's a Keynesian beauty contest. Each option resolves to its percentage at closing time.

@0xSMW Could you confirm what coproduct said? how will you count the votes and do you resolve YES to the one with the most votes and NO to all others or some other scheme

The top 1 will resolve to YES.

Yeah but how do you determine which one is the top 1

Are the votes from percentage odds, yes shareholder number, or separate poll?

determined via yes/no orders

As in just limit orders count? Is it number of users placing orders, number of total orders including multiple per user per option, amount of mana spent in orders, or number of shares in orders? Is it only fulfilled orders being counted?

I'm confused as to what is confusing?

if it resolved today, DeepSeek would resolve to yes, and all others no.

Is that based just on the percentage odds?

yes, it's very simple, it's based on the market itself – again, confused by the confusion?

Aren't you worried that resolving based only on odds will inventivise users with a lot of mana to non-epistemically pump up the odds of a random option right before market close i.e. whalebait?

no, I'm not. in fact, if it did, it would be the best thing that could happen to the market. it would draw more attention to the injustice and perceived truth.