korrents

On the map

Tap a claim on the ring to put it at the centre.

← Spending compute at inference time to search over plausible solutions…

8 connected korrents · 5 moments on record from 19 Jun 2024 to 31 Jul 2026.

Same subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subject Read this korrent: Spending compute at inference time to search over plausible solutions is the general fix for unreliable long agent runs. Spending compute at inference timeto search over plausible solutionsis the general fix for unreliablelong agent runs. JD Jeff Dean — holds since 2026-07-30 — tap for who they are Same subject: Inference, not training, is now the constraint, and specialized low-energy hardware will beat general-purpose GPUs and TPUs on latency. — tap to centre the map on it Inference, not training, isnow the constraint, andspecialized low-energyhardware will beat… JD Jeff Dean — holds since 2026-07-30 — tap for who they are Same subject: Real reasoning will have arrived when spending more compute at inference time reliably buys a dramatically better answer. — tap to centre the map on it Real reasoning will havearrived when spending morecompute at inference timereliably buys a dramatically… AS Aravind Srinivas — holds since 2024-06-19 — tap for who they are Same subject: Long-running agents fail because they drift off the distribution they were trained on, and degrade further the farther out they get. — tap to centre the map on it Long-running agents failbecause they drift off thedistribution they were trainedon, and degrade further the… JD Jeff Dean — holds since 2026-07-30 — tap for who they are Same subject: Agents can already be run for days or weeks on a single hard problem, and almost nobody has internalised that. — tap to centre the map on it Agents can already be run fordays or weeks on a single hardproblem, and almost nobody hasinternalised that. JD Jeff Dean — holds since 2026-07-30 — tap for who they are Same subject: Demand for inference may be growing exponentially while GPU production can only grow linearly, and the tightening happens where those two lines cross. — tap to centre the map on it Demand for inference may begrowing exponentially whileGPU production can only growlinearly, and the tightening… DR Dax Raad — holds since 2026-05-27 — tap for who they are Same subject: Orchestrating thousands of agents is a new form of test-time compute, a fourth lever alongside network size, training data and training flops. — tap to centre the map on it Orchestrating thousands ofagents is a new form oftest-time compute, a fourthlever alongside network size… BC Boris Cherny — holds since 2026-07-27 — tap for who they are Same subject: Inference is a high-margin business even for a middleman: renting GPUs at scale, some models carry an eighty per cent margin over their sticker price. — tap to centre the map on it Inference is a high-marginbusiness even for a middleman:renting GPUs at scale, somemodels carry an eighty per… DR Dax Raad — holds since 2026-05-27 — tap for who they are Same subject: Knowing something yourself will stay faster than asking a model for it, because a lookup in your own head beats a round trip to an agent. — tap to centre the map on it Knowing something yourselfwill stay faster than asking amodel for it, because a lookupin your own head beats a round… PC Patrick Collison — holds since 2026-07-31 — tap for who they are
same subject or similar wording

At the centre Spending compute at inference time to search over plausible solutions is the general fix for unreliable long agent runs. Holds Jeff Dean Read this korrent →