Trendoday

Ensemble Routing for Prompt-Adaptive Cost-Aware LLM Inference

Large language models (LLMs) increasingly support Web-based services that must balance response quality and inference cost across heterogeneous requests. Conventional routing selects one model per pro

  • Upvotes5
–Rank now
#12Peak rank
2Days on list
241Score · 7d
Nov 12, 2026First seen
Rank · 7 days

About this paper

PISA and PISA-E use prompt-adaptive capability estimation to route prompts or select cost-aware ensembles, improving cost-performance trade-offs across LLM routing benchmarks.

Large language models (LLMs) increasingly support Web-based services that must balance response quality and inference cost across heterogeneous requests. Conventional routing selects one model per prompt, while fixed ensembles cannot adapt their composition and cost to individual requests. We introduce PISA, a prompt-adaptive, cost-aware single-model router, and PISA-E, its extension to ensemble routing. At the core of both methods, PISA combines soft cluster memberships with model-specific predictors to estimate individual model capabilities. Building on these estimates, PISA-E jointly selects ensemble composition and cardinality under an explicit cost--quality preference that can be…

large language models (LLMs)prompt-adaptive routingcost-aware single-model routerensemble routingsoft cluster membershipsmodel-specific predictorsensemble compositionensemble cardinalitycost-quality preferenceensemble-success surrogate
Upvotes
5
Authors
4
Published
Nov 6, 2026

Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.

Trending history

Ensemble Routing for Prompt-Adaptive Cost-Aware LLM Inference was first seen on Hugging Face · Daily papers on Nov 12, 2026. Since then it has appeared in 20 checks on 2 different days. Its best position was #12 of 30, reached on Nov 13, 07:28 UTC. It dropped off the list after Nov 13, 21:28 UTC.

Rank history

Position on Hugging Face · Daily papers at every check (lower is better).

#1#11#20#30 07:2816:5802:2811:5821:28 Nov 12, 07:28 UTC · #26Nov 12, 09:28 UTC · #20Nov 12, 11:28 UTC · #18Nov 12, 13:28 UTC · #17Nov 12, 15:28 UTC · #15Nov 12, 17:28 UTC · #15Nov 12, 19:28 UTC · #15Nov 12, 21:28 UTC · #15Nov 12, 23:28 UTC · #15Nov 13, 01:28 UTC · #14Nov 13, 03:28 UTC · #14Nov 13, 05:28 UTC · #13Nov 13, 07:28 UTC · #12Nov 13, 09:28 UTC · #16Nov 13, 11:28 UTC · #15Nov 13, 13:28 UTC · #17Nov 13, 15:28 UTC · #17Nov 13, 17:28 UTC · #17Nov 13, 19:28 UTC · #17Nov 13, 21:28 UTC · #16

Last 14 days

Best rank per day (UTC).

Nov 3today

Metric history

Platform metric as reported by Hugging Face.

05914 07:2816:5802:2811:5821:28 Nov 12, 07:28 UTC · 5Nov 12, 09:28 UTC · 9Nov 12, 11:28 UTC · 10Nov 12, 13:28 UTC · 11Nov 12, 15:28 UTC · 12Nov 12, 17:28 UTC · 13Nov 12, 19:28 UTC · 14Nov 12, 21:28 UTC · 14Nov 12, 23:28 UTC · 14Nov 13, 01:28 UTC · 14Nov 13, 03:28 UTC · 14Nov 13, 05:28 UTC · 14Nov 13, 07:28 UTC · 14Nov 13, 09:28 UTC · 14Nov 13, 11:28 UTC · 14Nov 13, 13:28 UTC · 14Nov 13, 15:28 UTC · 14Nov 13, 17:28 UTC · 13Nov 13, 19:28 UTC · 13Nov 13, 21:28 UTC · 13

Questions

Is Ensemble Routing for Prompt-Adaptive Cost-Aware LLM Inference trending on HF Daily Papers right now?

Not at the moment. It was last on the list on Nov 13, 21:28 UTC, after 2 days of trending.

What is the highest rank Ensemble Routing for Prompt-Adaptive Cost-Aware LLM Inference reached?

#12 on Hugging Face · Daily papers, on Nov 13, 07:28 UTC. Trendoday checks this list every 120 minutes.

How long has Ensemble Routing for Prompt-Adaptive Cost-Aware LLM Inference been trending?

It was first seen on Nov 12, 2026 and has been on the list on 2 days (20 checks) since.