Stefano Ermon 1 item
LatestBriefsLong-formDiscussionsNewsCompaniesReviewsArchive
People Search
Long-form 2026-09-19 · 05:20 ET
No Priors — Ermon/Inception: diffusion wins inference parallelism
Stefano Ermon on Mercury ≈ Haiku/Flash/mini speed tier, ~10× decode vs AR at GPT-2 scale, OpenCall leaving Cerebras for NVDA GPUs, 20–30% latency wedge.
asr No Priors · Inception · Mercury · OpenAI
‹ All posts People /Stefano Ermon ARCHIVE
Person

Stefano Ermon

1 document in the archive.

Appearances newest first
No Priors — Ermon/Inception: diffusion wins inference parallelism
Stefano Ermon on Mercury ≈ Haiku/Flash/mini speed tier, ~10× decode vs AR at GPT-2 scale, OpenCall leaving Cerebras for NVDA GPUs, 20–30% latency wedge.
Sat, 19 Sept 2026 · 05:20 ET Long-form asr