How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip
Thread
Unofficial Hacker News client; not affiliated with Y Combinator.
How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip
Unofficial Hacker News client; not affiliated with Y Combinator.
pama · · focus · HN ↗
> When the first chips came back from the foundry in May, the team pointed its internal AI models at designing software to run benchmarks such as SemiAnalysis’s InferenceX. On DeepSeek’s multi-head latent attention kernel benchmark, performance climbed from 0.31 percent of the theoretical ceiling (set by the chip’s compute and memory bandwidth) to 88.94 percent in roughly 40 hours. Ho says this result is repeatable, so the time between when foundries deliver the first chips and when production ramps up can be reduced. “All our schedule assumptions are going to be based on the fact we have this capability now,” he says.
wmf · · focus · HN ↗
brookst · · focus · HN ↗
gregsadetsky · · focus · HN ↗
The tooling is open source, and Fable in a loop - especially when paired with a digital scope that Fable interfaces with (the Saleae’s [1] are great) - gives you a level of verifiability that feels like beyond what software typically gives you. ie it feels more like Lean than code with tests.
I had ai implement a few toy circuits (sha hashing, 8088 emulation, a tiny llm) but yeah. Still looking for fun applications.
There have been a few recent fpga threads on hn, check them out. [2][3]
[0] <a href="https://1bitsquared.com/products/icebreaker" rel="nofollow">https://1bitsquared.com/products/icebreaker
[1] <a href="https://www.saleae.com/" rel="nofollow">https://www.saleae.com/
[2] <a href="https://news.ycombinator.com/item?id=49564064">https://news.ycombinator.com/item?id=49564064
[3] <a href="https://news.ycombinator.com/item?id=49531525">https://news.ycombinator.com/item?id=49531525
brookst · · focus · HN ↗