BV-SALA Savings Layer: Verifiable Measurement of Cost Savings in Large Language Model Applications (Open-Core Edition)

Technical report, v2.3-r1 (15 November 2026), revised after an external pre-publication review. BV-SALA is an open-core SDK and gateway integrating established optimization mechanisms with usage accounting. The paper specifies provider usage as disjoint billable categories, distinguishes observed costs from reference-valued and estimated counterfactuals, and separates mechanism diagnostics from experimental attribution. Recomputed public September 2026 artifacts yield a pooled list-price cost reduction of 31.50% (percentile family bootstrap interval 25.74-36.75%) on gpt-4o-mini and 75.08% (69.31-79.60%) on claude-sonnet-5; the corresponding equal-family means are 22.88% and 70.71%. November runs (cacheable-prefix reorder, 0% to 95% hit-rate and -82.76% per request; dedup x cache_control factorial, -85.94%; cascade cost pilot, -67.91%) are presented as exploratory and do not establish quality non-inferiority. Review snapshot: commit 239c84b99fe8320f7efaa9da9e63def08d62aed6.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-16
DOI
https://doi.org/10.5281/zenodo.24278851
Primary Topic
Cloud Computing and Resource Management
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

BV-SALA Savings Layer: Verifiable Measurement of Cost Savings in Large Language Model Applications (Open-Core Edition)

Joan-Marc Fisa
4 links
Zenodo (CERN European Organization for Nuclear Research)
Cloud Computing and Resource Management
preprint

BV-SALA Savings Layer: Verifiable Measurement of Cost Savings in Large Language Model Applications (Open-Core Edition)

Joan-Marc Fisa
preprint en

Abstract

Technical report, v2.3-r1 (15 November 2026), revised after an external pre-publication review. BV-SALA is an open-core SDK and gateway integrating established optimization mechanisms with usage accounting. The paper specifies provider usage as disjoint billable categories, distinguishes observed costs from reference-valued and estimated counterfactuals, and separates mechanism diagnostics from experimental attribution. Recomputed public September 2026 artifacts yield a pooled list-price cost reduction of 31.50% (percentile family bootstrap interval 25.74-36.75%) on gpt-4o-mini and 75.08% (69.31-79.60%) on claude-sonnet-5; the corresponding equal-family means are 22.88% and 70.71%. November runs (cacheable-prefix reorder, 0% to 95% hit-rate and -82.76% per request; dedup x cache_control factorial, -85.94%; cascade cost pilot, -67.91%) are presented as exploratory and do not establish quality non-inferiority. Review snapshot: commit 239c84b99fe8320f7efaa9da9e63def08d62aed6.

Zenodo (CERN European Organization for Nuclear Research)
Cloud Computing and Resource Management
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon