The Model Index
Up to four models

Muse Wonder vs GPT-5.6 Terra

On the index (GPT-4 = 100), Muse Wonder scores 182 and GPT-5.6 Terra 178. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Muse Wonder
Presets
Muse Wonder
182166–199
WeightsClosed
ReleasedOct 2026
Price in / out$4.0 / $20
Context1.05M
Core tests taken1 of 9
GPT-5.6 Terra
178162–195
WeightsClosed
ReleasedJul 2026
Price in / out$2.0 / $12
Context1.05M
Core tests taken6 of 9

Muse Wonder and GPT-5.6 Terra are too close to call (4 points apart, within the uncertainty), ahead on 0 of 1 shared core tests, at 1.8× the price.

Test by testthird-party results

Muse WonderGPT-5.6 Terra
0%25%50%75%100%MATHSFrontierMath (tiers 1-3)FrontierMath (tiers 1-3)AIME-style maths (OTIS mock)AIME-style maths (OTIS mock)FrontierMath (tier 4)FrontierMath (tier 4)AGENTSDTBenchDTBenchFurniture assemblyFurniture assemblyLMCALMCAMystery game puzzlesMystery game puzzles
Every result as a table
TestMuse WonderGPT-5.6 Terra
Maths
FrontierMath (tiers 1-3)82.1%86.0%
AIME-style maths (OTIS mock)99.4%99.7%
FrontierMath (tier 4)70.7%70.7%
Agents
DTBench95.2%93.3%
Furniture assembly54.2%54.2%
LMCA56.8%54.9%
Mystery game puzzles51.0%35.0%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG