Meta AI·released 2026-09-021 source
Muse Spark 1.3 benchmark scores: 10 benchmarks tracked. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 10 benchmarks·best result 99.17 on OTIS Mock AIME 2024-2025 (unverified)·0 of 10 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 99.17accuracy (%) | unverifiedT2 | 2026-09-02 |
| DTBench | 94.22accuracy (%) | unverifiedT1 | 2026-09-02https://conceptualreasoning.ai/dtbench |
| FrontierMath-Tiers-1-3-v2-Private | 74.39accuracy (%) | unverifiedT2 | 2026-09-02 |
| LMCA | 63.38accuracy (%) | unverifiedT1 | 2026-09-02https://conceptualreasoning.ai/lmca |
| ProofBench | 58accuracy (%) | unverifiedT1 | 2026-09-02https://www.vals.ai/benchmarks/proof_bench |
| APEX-Agents | 57.8accuracy (%) | unverifiedT2 | 2026-09-02 |
| FrontierMath-Tier-4-v2-Private | 46.34accuracy (%) | unverifiedT2 | 2026-09-02 |
| Chess Puzzles | 33.71accuracy (%) | unverifiedT2 | 2026-09-02 |
| FrontierSWE | 25.8accuracy (%) | unverifiedT1 | 2026-09-02https://www.frontierswe.com/ |
| Mystery Game Puzzles | 17.38accuracy (%) | unverifiedT2 | 2026-09-02 |
Claims drawn from cited facts, not live model generation.
This model was developed by Meta AI and released in 2026.
2 cited facts
This model has scores on 10 tracked benchmarks and currently leads 0 of them. The record’s evidence status is unverified.
3 cited facts
Its mean gap to the best recorded score across its tracked benchmarks is 30.49 points.
1 cited fact
1 facts cross-checked across data sources: 1 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks
Muse Spark 1.3 is an AI model developed by Meta AI, released 2026-09-02. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Muse Spark 1.3 has recorded scores on 10 benchmarks, each shown with its evidence status.
Muse Spark 1.3 has recorded scores on 10 benchmarks — OTIS Mock AIME 2024-2025, DTBench, LMCA, Chess Puzzles, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private, and 4 more. The full table above shows each score with its evidence status.
0 of 10 recorded scores (0%) are independently reproduced rather than self-reported by the lab.
Muse Spark 1.3 has 10 tracked claims: 10 unverified.