Inference providers are not hosting Perplexity's decider model despite production demand
bo_wangbo · x · 2026-11-16
Multiple production customers are asking for Perplexity's decider model, but not enough inference providers host it, according to the post, which also praises the model's strong eval performance—highlighting a hosting supply gap for smaller vendors' models.
More from Infra
- Enterprise AI economics: cost the whole workflow, not just per-token pricing — kashifmanzoor · 2026-11-16
- llama.cpp PR brings ~1.3x Q2_0 prefill speedup on AVX-VNNI CPUs, big MoE gains too — pmttyji · 2026-11-16
- Engineer job-hunting with Pony, a Go runtime that supervises multiple AI coding agents — sharpeye_wnl · 2026-11-16
- NYT: Jensen Huang's year-long secret about Megaspeed ties led to his downfall at Nvidia — GaryMarcus · 2026-11-16
- Bindu Reddy: Nvidia's 80% margin unsustainable as Neocloud financing props up its own demand — bindureddy · 2026-11-15
- Jensen Huang personally courting candidates in quiet Nvidia CEO succession search — WaPo — Polymarket · 2026-11-15