tensor.news
Integrity IndexModelsCompareLabsPapersMethodologyAbout
tensor.news

© 2026

Evidence

  • Integrity Index
  • Model Claim Ledger
  • Compare

Entities

  • Models
  • Benchmarks
  • Labs
  • Papers

Method

  • Methodology
  • About
  • Privacy

Machine

  • Feed (Atom)
  • llms.txt
  • llms-full.txt
  • Sitemap

tensor.news grades benchmarks and tags model claims by evidence status. Every fact carries a source record and provenance. Scores are framed as task performance under a disclosed harness, never as deployed capability.

← Models

Nemotron-4 15B

NVIDIA·released 2024-02-27

self-reported

Scored on 7 benchmarks. 0% of its scores are independently reproduced — the rest are self-reported or unverified.

7
Benchmarks scored
0
SOTA leads
76.53
Best score
0%
Reproduced
0 of 7

Benchmark fingerprint

Every score, ranked, colored by evidence status.

  • HellaSwag
    76.53
  • PIQA
    64.8
  • Winogrande
    56
  • GSM8K
    46
  • MMLU
    44.93
  • BBH
    44.93
  • ARC AI2
    40.67

All benchmark scores

BenchmarkScoreEvidence statusMeasured
HellaSwag76.53accuracy (%)self-reported· optimized2024-02-27
PIQA64.8accuracy (%)self-reported· optimized2024-02-27
Winogrande56accuracy (%)self-reported· optimized2024-02-27
GSM8K46accuracy (%)self-reported· optimized2024-02-27
MMLU44.93accuracy (%)self-reported· optimized2024-02-27
BBH44.93accuracy (%)self-reported· optimized2024-02-27
ARC AI240.67accuracy (%)self-reported· optimized2024-02-27
7 source records · epoch_benchmarks. Every score is evidence-linked.

Follow the record

More from NVIDIA

  • Qwen3.6-35B-A3B-NVFP4
  • Gemma-4-26B-A4B-NVFP4
  • diffusiongemma-26B-A4B-it-NVFP4
  • Gemma-4-31B-IT-NVFP4
  • NVIDIA-Nemotron-3-Nano-4B-BF16
  • NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4