MetaUnited States· released Apr 2026

Muse Spark

Name: Muse Spark
Author: Meta

Provisional: 8 comparable tests so far. Ranked once it reaches 6 across 3 counting categories.

Meta, United States. One of 4 models from this lab on the Index.

Third-party publishedLow confidence (thin coverage)

Snapshot September 8, 2026

OpenCharts Index

71.3

Provisional: not enough comparable tests or counting categories to rank yet · 8 comparable tests · 1 counting categories

Drop any single test and the Index lands between 71.3 and 82.9.

Needs more results (1 of 8 so far)

Averaging 87.2 on the 1 it has; the score opens at 4 of 8.

Distance to today's frontier on the hardest tests. Not a measure of intelligence.

What it's known for

Muse Spark, in 8 results.

Strongest in reasoning (71.3). Its widest gap is on ProofBench, 83.0% behind the best published result. No comparable results yet in knowledge, agentic and long context. Between 25% and 30% behind the best published results, on average.

Strongest category

Reasoning

Hard, novel problems: graduate science, abstract puzzles, expert exams.

71.3

GPQA Diamond89.8% · 92% of best
Humanity's Last Exam40.6% · 87% of best
CritPt11.3% · 35% of best

Mean of 3 of the category's 6 comparable tests. Counts toward the Index.

Best single result

Text Arena

Answer the same prompt as a rival model; real users vote blind on which answer they prefer.

1,488

47% expected win rate against the board leader (1,507) · #11 of 92.

Millions of blind votes on real prompts are the market's measure of what people prefer.

Where it trails

ProofBench

Write a complete, rigorous proof, graded for rigor rather than for a final answer.

17.0%

83.0% behind the best published result (100.0%) · #40 of 61.

A proof shows the reasoning itself. A right answer with a wrong argument scores nothing.

Not measured yet

No comparable result yet in these 3 categories. A blank is a blank, never a zero, and it does not lower the Index.

Knowledge
Agentic
Long context

Covered, but not averaged into the Index:

Coding1 of 7 tests · needs 4 · mean 83.0
Math2 of 5 tests · needs 3 · mean 53.0
Multimodalsingle-test category, shown beside the Index · 94.4
Human preferencesingle-test category, shown beside the Index · 94.5

Every score, with its source

See it at work

What Muse Spark was asked to do, and how it did.

Every test it has taken, by category, with the kind of task it faced, who produced the number and where the result landed against the best published one. The examples are original and illustrative, never items from the datasets themselves.

Human preference · mean 94.5 across 1 comparable test · a single-test category, shown beside the Index and never averaged into it

Text ArenaLeaderboard

Answer the same prompt as a rival model; real users vote blind on which answer they prefer.

1,488

47% expected win rate against the board leader (1,507)
#11 of 92

board 2026-09-02

Source: Arena

Every result is read against the best published one on its test: percent scores as a share of the best above chance, Arena ratings as the expected win rate against the board leader, open-ended values on a log scale. Sitting a harder exam never lowers a model. Every number links to the publisher that produced it.

Head to head

Muse Spark against whoever you pick.

Choose a rival. Every counted test both have taken appears side by side as a share of the best published result on that test, with the category means above. A win is a higher share; a gap under 2 points is a tie, because one item on a 45-problem exam is 2.2 points and publishers' own standard errors are of that order.

Muse Spark

provisional·Index 71.3B-

0 : 5

wins · 0 ties · 5 shared

GPT-6 Astra

#1·Index 99.0A+

GPT-6 Astra comes out ahead on 5 of the 5 tests both have taken.

Category means

Reasoning
71.3
99.6
Knowledge
—
100.0
Coding
83.0
97.7
Math
53.0
99.8
Agentic
—
98.5
Multimodal
94.4
—
Human preference
94.5
—

Shared tests · biggest gaps first

ProofBenchMath
17.0%
99.0%
−82.0
CritPtReasoning
11.3%
31.7%
−63.1
Mock AIME 2024–2025Math
88.9%
100.0%
−11.1
GPQA DiamondReasoning
89.8%
95.8%
−8.4
SciCodeCoding
51.5%
56.5%
−8.0

Bars are distances to the best published result on each test (percent scores above chance, Arena ratings as win rate against the leader, open-ended values on a log scale). Raw values as the publishers report them.

Open GPT-6 Astra

Scores

Every test, every source

The best published run per benchmark, the raw value as the publisher reports it, who produced the number, how close it comes to the best published result on that test, and where the model sits among every model scored on it. Percent scores are read above chance, Arena ratings as a win rate against the board leader, open-ended values on a log scale. A test with too few models to compare against is shown but not counted.

Category means

Reasoning71.3 · 3/6
Knowledge—
Coding83.0 · 1/7
Math53.0 · 2/5
Agentic—
Multimodalbeside94.4 · 1/1
Human preferencebeside94.5 · 1/1
Long context—

Mean · tests taken of the category's comparable tests. A faded category either rests on fewer than half of them or is a single-test category (marked “beside”), which is shown next to the Index and never averaged into it.

GPQA DiamondEpoch-runReasoning

89.8%

91.6% of the best (95.8%), above chance
60th percentile of 78 · n = 198

Source: Epoch AI — AI Benchmarking Hub

Humanity's Last ExamLeaderboardReasoning

40.6%

87.2% of the best (46.5%)
82th percentile of 18 · n = 2,500

Source: Epoch AI — AI Benchmarking Hub

CritPtLeaderboardReasoning

11.3%

35.1% of the best (32.3%)
58th percentile of 78

Source: Epoch AI — AI Benchmarking Hub