Absolutely. LLM inference is still a greenfield — things like overlap scheduling and JIT CUDA kernels are very recent. We’re just getting started optimizing for modern LLM architectures, so cost/perf will keep improving fast.
As a user of a lot of coding tokens I’m most interested in latency - these numbers are presumably for heavily batched workloads. I dearly wish Claude had a cerebras endpoint.
I’m sure I’d use more tokens because I’d get more revs, but I don’t think token usage would increase linearly with speed: I need time to think about what I want to and what’s happened or is proposed. But I feel like I would be able to stay in flow state if the responses were faster, and that’s super appealing.
https://hex.pm/packages/vllm
Makes you think that you will continue to see the costs for a fixed level of "intelligence" dropping.
(PyTorch does also support ROCm generally, it shows up as a CUDA device.)
I’m sure I’d use more tokens because I’d get more revs, but I don’t think token usage would increase linearly with speed: I need time to think about what I want to and what’s happened or is proposed. But I feel like I would be able to stay in flow state if the responses were faster, and that’s super appealing.