6 ms·
Because it's not losing money on each token? Aside from most global people using American inference providers to run the models, I suspect the cloud inference p
by m11a 2mo ago
Because it's not losing money on each token? Aside from most global people using American inference providers to run the models, I suspect the cloud inference products of the Chinese labs are profitable, at least on the inference costs (ie: not including model training, salaries, etc).
- georgeburdell 2mo agoThere is no way Deepseek is making money even on inference
- arczyx 2mo agoPretty sure they are making money since on OpenRouter, there are other providers for DeepSeek V4 flash that are charging even less than DeepSeek themselves (eg DeepInfra and Digital Ocean). https://openrouter.ai/compare/deepseek/deepseek-v4-flash/tencent/hy3-preview/google/gemini-2.5-flash-lite https://openrouter.ai/compare/deepseek/deepseek-v4-flash/ten...
- culi 2mo agoThat doesn't mean they're making money. Other providers could just be doing it more efficiently. DeepSeek is trying to make its model optimized for Hauwei chips instead of NVIDIA so it has its own constraints
- georgeburdell 2mo agoThat says little. Those providers could also be losing money trying to gain marketshare. There is a high amount of speculation in the space and it won’t be apparent for awhile who has a lasting business.
- georgeburdell 1mo agoAnd here we go: https://news.ycombinator.com/item?id=49192693 https://news.ycombinator.com/item?id=49192693