gizmodo59

Side note.. Fable just rejected this. GLM 5.3 did without questioning me. 5.6 sol did it beautifully.

I somehow find it better to give 2 frontier model companies 100-200/month than dropping 10 grand on a hardware that will get old in no time with bad TPS. I really want to have a fully local model but seems like one more generation wait and we will be there?

I will believe there is no moat when the revenues for Anthropic is not 70B. It seems like people want to throw away money and they don’t like switching

It will be comparable to Luna then.

how does this compare with https://developers.openai.com/api/docs/models/omni-moderatio...

As for use cases, obviously we can't fully rely on non-deterministic capability for sensitive things but a small model which can do a good job acts as a first defense and then a human can review later.

If your comment is referring to situational awareness, its due to 4x leverage. leverage is always risky. AI/semis are still doing extremely well (over last 2 years) despite the recent dip

Also: https://www.cnet.com/tech/tech-industry/apple-google-others-...

The web and connecting to other services is very important for almost all of my use cases. While I believe we are going to get better and faster models, the web index is certainly not downloadable and maintainable for 99.99% of the folks who are able to use local models. Any good solutions exist?

Moat is not the harness. Harness itself is temporary until the models get better and slowly the code in harness will go down.

Note that the biggest GPU providers in the world are the hyper scalers and even they couldn’t allocate more if you pay for it. Because the rich companies and well funded ones are gobbling them up to the point where if tomorrow a 5T model that smokes every other model in the world is released you just can’t afford inference.

That’s a very narrow view. So non of the fields of science and engineering matter but just cancer?

It’s also very very divided (x companies, oss vs not and other interests)

People still believe that the model intelligence has plateaud and I feel like they live under the rocks. I get the insane capex spend, I get the hype, I even get certain company will go under but to ignore the capability jump in a year is absurd.

Yet another "benchmark to promote their own harness"

When would I use this over the plugin in codex? Which I think can be invoked from cli as well

Because lack of talent and organizational disfunction matters a lot more than you think. The reason why OAI and Ant are always at the top is because of this and I’d say compute is third on the list.

I included Google along with Ant and OAI. OSS models have never been the highest performing model unless you refer to benchmaxxing.

Yes. I’m 99% sure arc agi 3 will be saturated like 1 and 2. In less than a year. And they will come up with one more.

None of the people signed this have ever produced a frontier model at a given date (Which is to say its neither Google/OAI/Ant). The ones that sign are meta, musk (who is in the shovel selling business as well), hugging face obviously and few others

Note: Just pointing out the comment intent and nothing else

what tool does someone recommend for claude and gemini? Its time to extract data from each!

A lot of things are not black and white and it’s nuanced so someone with higher intelligence will point it out. It’s an emergent behavior not a defect

MY cynical take: Until the compute needs get so enormous that only governments can fund it and there is a consensus internationally, its either company A in country X or company B in country Y. And since everyone thinks THEY are the good guys the competition will continue.

that looks exactly like the recent codex!

ultra in codex is just a way to run multi agent system, pro is similar to other pro models like 5.5

This number is obviously absurd but for other normal amount (say 2 million which can definitely happen with a mistake), can't they claim it in court that will bankrupt you even if your entire worth is just 100K for example?

responses api provided many benefits over chat completions https://developers.openai.com/api/docs/guides/migrate-to-res.... Any reasoning model which is all the new models these days should use responses as per the recommendation

“ Eyes and brain alike evolved over millennia to process natural scenes, forests, rivers, coastlines, open skies. These environments share a specific mathematical pattern: their visual complexity decreases predictably as you zoom in on finer and finer details.”

Doesn’t it increase the details in nature due to all the imperfections vs simple patterns made by humans?

Anthropic is getting extremely petty and especially against oai

- ad in superbowl about how they are the good guys.

- dow public PR stunt (they are the ones to give Palantir their model access).

- sues openclaw.

- threatens every use of cc in oss community.

- prevents other companies using claude saying they cant use when they compete.

- never released a single open weight model.

- Dario told OAI is Yolo'ing in compute and they are now doing the same.

- gas lighting developers and then after weeks acknowledging they fiddled with reasoning juice.

- fear mongoring on mythos and then geting compute later and acknowledging publicly once they realized its not significantly better than gpt 5.5 cyber.

- signs a deal with Elon!

- now this!

Inference is profitable. Companies lose money because:

1. Training is expensive. Not just compute but getting the data, researchers salaries etc 2. You have to keep producing new models to ensure people use your inference and there seems to be no end to this. So they have to pour more billions to keep the cycle going on 3. People salary and other admin cost are not that high compared to 1 and 2.

where are all the claude fans who hate elon? Its very funny how ant will do anything at this point.

The negative reactions here are baffling me. The fact that we can even get to say 30% with computer is amazing. So much hatred towards AI and anything from the frontier labs like OpenAI (or Goog for that matter) makes no sense.