[markasoftware]

Multiple providers (who need to make a profit) offer the same 4.40 rate for glm-5.2. It's not subsidized.

Deepseek's 0.86 or whatever is likely subsidized but alternate providers offer it for a price comparable to glm-5.2.

[throwa356262]

According to deepseek themselves, their current rates are NOT subsidised.

They have published tons of articles dedicated to performance and efficiency engineering. Feel free to have a look...

[markasoftware]

Why is no other inference provider offering similar prices then?

[throwa356262]

How long did it take vLLM to implement deepseeks sparse attention from the r1 paper?

Does ananyone outside deepseek have a working code for the v4 compressed attention mechanism?

Has any other provider managed to bypass CUDA and program the compute engines in their native assembly language to get 10% more performance out of them?

There is your answer.

[est31]

GPU/RAM/etc prices could continue to rise. If the world leaders decide it's time to build the robot armies, then that could price out the civilian uses for GPUs.