Yes, but DeepSeek went absolute goblin-mode on optimization and I would not be shocked to discover that OpenAI and Anthropic were behind on that front.

What is the TPS of DeepSeek? All of the LLMs I have tried are slow as molasses. \

I pretty reliably get 200 tokens per second on deep infra.

Deepinfra has a "contact sales" link. I assume personal accounts aren't available.

I use TensorX (EU hosted) and get 330TPS

OpenAI and Anthropic have more compute and the best frontier models to throw at the optimization problems. They have published articles on these efforts. For example, Dots, and Anthropic's upcoming "persistent agent" is because of these optimizations. Both companies often publish articles detailing a lot of their technical efforts. They just get buried among the flood of other news.

That's just it: they have more compute. They have a "do more with more" lever in addition to a "do slightly less with much less" lever. DeepSeek can't win with the first lever, so they have committed everything to the second, while within the frontier labs the two levers fight for attention and frankly the first is clearly the favored child. I'm not accusing OpenAI and Anthropic of being slouches, but I am not sure that more resources + divided goals > singleminded pursuit of a singular goal.

Yes, I would not be shocked either.

Sonnet 5.5 is 16x more expensive at $10 per million output tokens. Kimi K3.1 is also $10 per million output tokens, and its number of active parameters is only 6.5x larger than V4.1 Flash (104B compared to 16B). These things are not directly comparable of course, but just to give an idea.

Even if inference of those models needs 16x, 32x, or 48x the number of GPUs compared to V4.1 Flash, still seems like potentially great revenue per GPU.

My friend you are looking at the wrong number.

Its cash flows to the GPU that matter. Not revenue.

They published all/most of their work and it would be unbelievable if Anthropic had not already used that research in their own architecture.

Given that they open sourced their inference framework, how realistic would be to buy hardware and become a provider on OpenRouter? These numbers are incredible.

Does Deepseek say how much those Clusters cost?

Don't forget that Deepseek pays salary in RMB which is 1/7 of the value of USD.

It's sad how far down you have to scroll to get the real answer. It's because Anthropic margins on inference are absolutely bonkers and it's easy for them to drop prices by 80% and STILL have positive free cash flow

Missed the memo. Whats the proof of positive free cash flow?

It's all based on vibes and whatever vague lies the companies put out about their financials. Marvels of corpo-nonsense speak such as "If we ignored every single cost we have, we're quite profitable!" seems to be very popular with the AI crowd

If you need to strawman to make a point, you're probably not making a good point.

"We're profitable on inference. If we didn't pay for training, we'd be a very profitable company." - https://simonwillison.net/2025/Aug/17/sam-altman/

Where exactly is my strawman? Scam Altman himself has said this, if we ignore the inconvenient parts of the business that loses all that money, they'd be super-duper profitable.

Anthropic said something similar, claiming to their investors they're profitable... If you ignore partner costs and the costs of training.

And mind you this is coming from a proven pathological liar who has a massive vested interest in over exaggerating his business' viability, so doubtless the "Very profitable" part is itself a massive lie.

Gross profitability always happens before net profitability, lol. "We've reached the first but not the second" isn't a deceptive claim, especially with the caveat in the same sentence, and it's a step that every net profitable company went through at some point in its life.

The correct quote is "If you ignore the costs that will come down in the future or stay fixed as userbase scales, then we are actually quite profitable"

Which is a basic truism of pretty much all startups, and nothing noteworthy or insightful to point out.

Why would you expect training costs to come down in the future? It seems to me that the labs themselves have concluded they can't stop training due to competition, which is why they're now asking the government to step in and shut down their competitors.

> or stay fixed as userbase scales

This sentiment aged extremely poorly with Anthropic's S1 now released. Turns out the "vibes" guys were right! I expect the goalpost will shift to claims that the S1 is fabricated in a larger than Enron fraud involving both the auditors and Anthropic.

You mean the S-1 docs where they state they're operating at a GAAP of -4% (even then that's based on their best 28 days multiplied by 13), meaning losing money? You sure showed the plebs with the impressive billions of dollars of loss!

https://www.sec.gov/Archives/edgar/data/2130357/000162828026...

Operating cash flow of $3,929M and $482M of CapEx gives $3,447M of Free Cash Flow in H1 2026

I believe it was Eric Schmidt at Google who once famously said "revenue solves all known problems" in an all hands

There's a reason he didn't say profit...

Not saying I agree with the sentiment, just noteworthy and perhaps something to reflect upon while judging behaviors aligned with this perspective.

it doesn't exist. Anthropic overtook OAI as the most unprofitable company in human history

The S1 was filed yesterday. https://www.sec.gov/Archives/edgar/data/2130357/000162828026...

Operating cash flow of $3,929M and $482M of CapEx gives $3,447M of Free Cash Flow in H1 2026

-$70Bn

Leaked s1 filing, I would guess

https://www.sec.gov/Archives/edgar/data/2130357/000162828026...

Operating cash flow of $3,929M and $482M of CapEx gives $3,447M of Free Cash Flow in H1 2026

Their recently released S-1, though I don't think they do a detailed breakdown.

They state their operating cash flow and their capex. FCF is defined as OCF-Capex

The S1 that was just filed.

https://www.sec.gov/Archives/edgar/data/2130357/000162828026...

Operating cash flow of $3,929M and $482M of CapEx gives $3,447M of Free Cash Flow in H1 2026

How many times are you going to post this exact same comment in this thread?

I thought you were exaggerating but man…

> $38 of revenue per GPU

its per GPU-cluster, not a single gpu.

Based on translating their Chinese text on Firefox, it says "per card":

> The current service throughput of a single cluster is 155 trillion tokens per day, and the overall revenue is $38 per hour (at the lowest price) per card.

With Kagi translate (best), it says GPU

> A single cluster currently serves 155 trillion tokens per day, generating overall revenue of $38 per GPU per hour (at off-peak electricity rates).

https://zhuanlan.zhihu.com/p/2095560295930508953

that's significantly over price of buying and powering up the GPU

> its per GPU-cluster, not a single gpu.

No, it isn't. It's revenue per hour per GPU.

How can their revenue numbers match benchmarks from their inferencing code? Are they really claiming that they sell 100% of their capacity at all times, 24/7? Wouldn't they require a lot of queued batch traffic to achieve that? Or are they defining revenue in some weird way where what they mean is potential revenue?

I assume this is idealized somehow, given how close it is to DeepInfer numbers at 100% utilization. My point was more that there is seemingly a lot of margin to absorb underutilization, outages and other costs.

Is that revenue or profit?