Pith Number

pith:N5F6BWHH

pith:2026:N5F6BWHHJFCCKPWYLOFYBW3SUE

not attested not anchored not stored refs resolved

EnergyLens: Predictive Energy-Aware Exploration for Multi-GPU LLM Inference Optimization

Anantha P. Chandrakasan, Eun Kyung Lee, Kyungmi Lee, Tamar Eilam, Xin Zhang, Zhiye Song

EnergyLens predicts multi-GPU LLM inference energy with 9-13 percent error to identify efficient configurations without exhaustive profiling.

arxiv:2605.14249 v1 · 2026-05-14 · cs.LG

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{N5F6BWHHJFCCKPWYLOFYBW3SUE}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

EnergyLens achieves mean absolute percentage errors (MAPEs) between 9.25% and 13.19% for multi-GPU prefill and decode energy, and correctly identifies Pareto-optimal overlap configurations.

C2weakest assumption

The empirically driven communication energy model and load-imbalance-aware MoE modeling generalize accurately to unseen multi-GPU configurations and model scales beyond the validation set.

C3one line summary

EnergyLens predicts multi-GPU LLM inference energy consumption with 9-13% MAPE and identifies configurations with up to 52x energy efficiency differences.

References

22 extracted · 22 resolved · 5 Pith anchors

[1] El-rec: Efficient large-scale recommendation model training via tensor-train embedding table 2022 · doi:10.1109/sc41404.2022.00051

[2] Enhanced system- level coherence for heterogeneous unified memory architectures 2024 · doi:10.1109/iiswc63097.2024.00012

[3] arXiv:2602.23036 [cs]

[4] arXiv: 2309.14393

[5] arXiv: 2002.05651 2002

Receipt and verification

First computed	2026-05-17T23:39:10.580884Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

6f4be0d8e74944253ed85b8b80db72a12cf1a2d99e61c8376924e8dc306d10ba

Aliases

arxiv: 2605.14249 · arxiv_version: 2605.14249v1 · doi: 10.48550/arxiv.2605.14249 · pith_short_12: N5F6BWHHJFCC · pith_short_16: N5F6BWHHJFCCKPWY · pith_short_8: N5F6BWHH

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/N5F6BWHHJFCCKPWYLOFYBW3SUE \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 6f4be0d8e74944253ed85b8b80db72a12cf1a2d99e61c8376924e8dc306d10ba

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "5bbc917ec34735ad54347247b668c5c6a5741205b3ad0db9bd7330586fe7a3d3",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by-nc-nd/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-14T01:37:26Z",
    "title_canon_sha256": "6e0858dcb83a95e96faef8c9ab26efce86e89ea899aef03d9af9124e658bc503"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.14249",
    "kind": "arxiv",
    "version": 1
  }
}