Restraint: The AI Opportunity Is Too Big to Hoard

We really are just a group of ordinary people. If there's a narrative I like, it's that a group of ordinary people did extraordinary things - not that a group of geniuses did extraordinary things. This is closely related to our restraint, and it's consistent with our restraint and our vision.

So, is there a conflict between open source and commercialization? I think in AI, if you're not restrained, you won't succeed. Open source is part of being restrained, and our restraint is not only reflected in open source but also in many other aspects. But overall, we don't need to think about open source or restraint.

The more restrained you are, the more likely you are to succeed - or at least that's what has been validated so far, and it's explainable. Otherwise, there's no way to explain how we succeeded: we had no weapons, a very low starting point, very few resources, and our people are just a random group of ordinary people. I myself am just a college graduate, and not from the most top-tier school.

This restraint is also part of our vision. AI is too big, the benefits are too huge. We are very restrained; as long as we can succeed, the benefits will be enormous in the end. If you take just a small piece, the benefits are already huge, so there's no need to think about which part to take or how to take it. I think we don't need to consider it at all, because the benefits are big enough. Just taking a tiny fraction is more than enough.

So we said before, we only take a reasonable profit - it's about your intention, not about maximizing profit. That's different. This is not our API pricing. Our API pricing considers a reasonable profit: roughly, if we go to the market and buy a batch of equipment, we want to recover the cost in ten months. We think that's a reasonable profit. Under current circumstances, considering risks and upfront investments, if we amortize a server financially over three or five years, but commercially we think recovering cost in ten months is enough. OK, we think it's enough. So that's the logic behind our current API pricing. Our V3.2 Flash and others all recover equipment cost in ten months.

That's our standard. It's actually not profit-maximizing. If we wanted to maximize profit, we would set the price higher. Because in this price range, user demand is inelastic: even if I double the price, the token consumption doesn't change much. If I double the price, my total revenue nearly doubles. Wait, let me check. Oh, that's great.

Let me tell you a story about our DDCP model. At first, we worried about too much demand, so we set the price relatively high. The team wasn't very happy about that. Later, I lowered the price to a quarter, and everyone was very happy.

I think this is our real intention. It's the vision I mentioned earlier: we want this to be useful to people, not to make the most money, but to make it affordable for everyone while earning a reasonable profit. I think that's what others in the company think too - when we lowered the price, many people in the company group chat cheered, everyone was very happy.

Because this is the purpose of all the effort and care we put into making this model good. The goal is to make it very cheap, very effective, and let everyone use it to the fullest. That makes us happy. That's our motivation, our vision, the consensus that brings our company together to do this. That's our internal consensus.

This should be quite unique, because lowering prices is definitely not good for our competitors. They certainly wouldn't cheer. Because your revenue, your ARR - if you cut it in half, ARR drops by half. Yes, that's a difference with us.

We think that's enough. Internally, if we recover cost in ten months, I'm already very satisfied commercially. Externally, we think this price is something everyone is happy and willing to see. It's a win-win for everyone: the company, society, and everyone.

I think, OK, someone just left a comment on the screen saying that breaking even in ten months means the profit is too high. There is indeed room to lower prices. There is also room for model optimization, so overall there is significant room to reduce prices.

But achieving cost recovery in ten months - we can do it ourselves, but others can't. Like Alibaba or Tencent, they don't have our optimizations, so their costs are probably several times higher. There's still a lot of optimization work to be done.

Earlier I mentioned why we don't lower prices further: it's because demand is inelastic. If I lower prices further, demand won't increase much, or will increase very little. Because at this price, everyone can afford it, and everyone is satisfied with the price. They won't stop using it because it's too expensive.

So lowering prices first of all won't bring the company more revenue, and it won't bring more value to society either, because everyone is already satisfied with this price. If you make it even cheaper, it won't increase society's happiness much. Yes, OK, but on this issue, in terms of pricing, we certainly are not aiming to have the highest revenue or highest profit as our starting point. That's part of our restraint. Because in the short term, if you price higher, you might earn more; but in the long run, it's hard to say. I think restraint is a strategy. For me, restraint is a strategy. It means sometimes you can give up something to gain more other things. The same goes for not open-sourcing. It can be seen as our pressure or our concession.