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

Muse Spark

Meta AI·released 2026-04-08

independently reproduced

Scored on 7 benchmarks, leading 1. 71% of its scores are independently reproduced — the rest are self-reported or unverified.

7
Benchmarks scored
1
SOTA leads
88.89
Best score
71%
Reproduced
5 of 7

Leads on: FrontierMath-2025-02-28-Private

Benchmark fingerprint

Every score, ranked, colored by evidence status.

  • OTIS Mock AIME 2024-2025
    88.89
  • GPQA diamond
    86.4
  • FrontierMath-2025-02-28-Private
    68.42
  • SimpleQA Verified
    66.3
  • HLE
    37.56
  • FrontierMath-Tier-4-2025-07-01-Private
    24.33
  • CritPt
    11.33

All benchmark scores

BenchmarkScoreEvidence statusMeasured
OTIS Mock AIME 2024-202588.89accuracy (%)reproduced· optimized2026-04-08
GPQA diamond86.4accuracy (%)reproduced· optimized2026-04-08
FrontierMath-2025-02-28-Private68.42accuracy (%)reproduced· optimized2026-04-08
SimpleQA Verified66.3accuracy (%)reproduced2026-04-08
HLE37.56accuracy (%)unverified2026-04-08
FrontierMath-Tier-4-2025-07-01-Private24.33accuracy (%)reproduced2026-04-08
CritPt11.33accuracy (%)unverified2026-04-08
7 source records · epoch_benchmarks. Every score is evidence-linked.

Follow the record

More from Meta AI

  • Llama 3.1-405B93.73
  • Llama 2-70B87.6
  • LLaMA-65B86
  • Llama 2-34B84.6
  • LLaMA-33B83.8
  • Llama 3.1-8B82.4

Benchmarks it leads

  • FrontierMath-2025-02-28-Private