system2

I am using the API only with them for now. But what you describe is nothing compared to Qwen or Mimo. These models are more capable than Opus in general and at a fraction of Opus's cost.

No. I got more done with Opus 4.8 than with Opus 5.5. It is at least 10x slower than Opus 4.8. A task that would take 3 minutes now takes 1 hour and is full of mistakes.

Wowza. The placeholder text for search says "Help me find a new job". Couldn't they find something less depressing?

If sony released jailbroken ps5 I'd buy it in a heartbeat. Unsure why they do not let us use it as much as we can. They would definitely sell more.

Chinese model pressure. Many of my SWE friends switched to Chinese models. I also use QWEN and GLM for many of the api requiring projects and dropped OpenAI and Anthropic. The only reason was the cost.

EDIT: I love getting downvoted by openai and anthropic employees or their bots.

They are just not overcharging. Nvidia's top AI chip Rubin sells in 72-GPU racks for about $3.5–7.8M. A rack running Xiaomi's MiMo V2.6 Pro generates roughly 150–300B tokens a day, worth about $130–260k at Xiaomi's API price.

That's a payback of the infrastructure in a few weeks in theory. After a few weeks or a month, the only cost is electricity, and whatever they make after that is pure profit. This is why they can charge normal prices. Not $50 for 1M tokens.

Once again, just like your other reply, you are mixing things up- a logical fallacy called the straw man; look it up. Your responses usually use this type of misdirection. Coding agent =/= API.

AI does not mean coders coding with agents all day long. AI integration is mostly for data processing, which is the promise and use of the APIs.

They are not the ones spending API money.

Most businesses I know switched to Google Sheets or Google Docs.

Not him but 3 models dominate: GLM 5.3, Qwen 3.8, Mimo 2.6. All censoring certain things. Numbers and other uses are perfectly fine. They are like 0.10-0.15 per 1M tokens. American AI lost the game already, people just can't see it.

Make 1M tokens $0.10; then I will use Sonnet. Until then, it is garbage.

I listened to This American Life from the first episode until they abandoned their mobile app. It was so easy for me to use it and mark things. I even emailed them about it; they said it cost too much to operate. It didn't make sense to me. I guess I will go back to listening to them again on Apple Podcasts. I even exported my playlist with data from their old app; not sure how I will sync with the Apple Podcasts app, though... Another challenge for Fable to solve, I guess.

A super sloppy vibe coded theme. That's why.

I guess someone will need to clone this site with a search function, categorization, and proper pagination. Nice attempt, though.

I vividly remember the agony of not having an AGP port on my motherboard and being forced to buy Voodoo 3-2000 (PCI). It was not as good as the Voodoo 3-3000, which was AGP, but finally I was able to play Quake2 with 3d acceleration. Good memories.

We are using it with OpenAI Luna. We send any failed query to Luna, and the operation is complete.

No wonder why they are all radicalized. 123 kids. Jesus.

News mostly. Anything China-related gets censored without hesitation. Some random stuff got censored too. It is borderline unusable, to be honest, unless only numbers are crunched.

Funny I was duped into buying Cyrix instead of Pentium 2. The guy was telling me it would play Quake better. It didn't, of course.

I think large database-related projects. Ai context will never be billions of tokens. And prostitution on the side. With both, we will make a good living.

Except for the censorship. We use it for massive data crunching, and roughly 5-8% (depending on the day) gets censored and doesn't get a response. We switched to Mimo 2.6, which is relatively better. For censored stuff, we use Sonnet and OpenAI Nano models.

Also Mimo 2.6 is roughly 30% cheaper. Without batch.

Not a source but a comparison with a weaker non-SOTA model:

Nvidia's top AI chip Rubin sells in 72-GPU racks for about $3.5–7.8M. A rack running Xiaomi's MiMo V2.6 Pro generates roughly 150–300B tokens a day, worth about $130–260k at Xiaomi's API price. That's a payback of the infrastructure in a few weeks in theory. After a few weeks or a month, the only cost is electricity, and whatever they make after that is pure profit.

OpenAI and Anthropic are practically scamming people with the token prices.

Or come up with good projects that utilize these and provide services that Ai alone gannot provide.

pearcmd.php must exist, and register_argc_argv must be on, not common with hosting providers. But I am assuming the other themes and meeting conditions possibly affect a lot of WordPress sites.

Nvidia's top AI chip, Rubin, sells in 72-GPU racks for about $3.5–7.8M. A rack running Xiaomi's MiMo V2.6 Pro could generate roughly 150–300B tokens a day, worth about $130–260k at Xiaomi's API price. That's a payback of a few weeks in theory.

So yes, we are getting scammed by American SOTA.

All I care about is the token price for the API. Haiku cannot get close to GLM or Mimo.

Everyone is collectively hating on Anthropic, so this news is just adding to the fun.

At least after buying 166mmx you didn't get lowered to 133mhz after a month. The progress was legit back then.

That's not chinese models' fault. Fable and Astra deserve to be punished for their scammy bait-and-switch.

It sounds like they are pricing very well because OpenAI and Anthropic have been scamming us for years. Once the infrastructure is in place, electricity cost is the only concern. China supports businesses and gives them a lot of incentives to lower their costs. And who knows what's being provided to them without anyone knowing. All in the name of winning the race.