SShogo Bench

GPT-5.6 Terra vs Muse Spark 1.3

GPT-5.6 Terra leads Muse Spark 1.3 by 4.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-5.6 Terra wins 7 and Muse Spark 1.3 wins 7 of 14 shared benchmarks. Muse Spark 1.3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGPT-5.6 TerraMuse Spark 1.3
Agentic coding6660
Long-horizon autonomy6964
Knowledge-work agents4745
Computer use4843
Security & context learning5653

Shared benchmarks

BenchmarkGPT-5.6 TerraMuse Spark 1.3
APEX-Agentsindex58.2%58.6%
CursorBenchindex41.3%41.6%
Chess Puzzles47.0%37.0%
CritPt30.0%26.0%
DTBench93.3%96.5%
FrontierMath Tier 470.7%46.3%
FrontierMath Tiers 1-386.0%74.4%
LMCA54.9%53.9%
MirrorCode41.1%54.4%
OTIS Mock AIME99.7%99.2%
ProofBench74.0%58.0%
SciCode55.0%59.7%
SimpleBench48.9%81.8%
WebDev Arena15211652

Specs and price

GPT-5.6 TerraMuse Spark 1.3
Agent Index58.8 (#16)54.7 (#28)
Agent $/M$1.02$0.030
Input / output $/M$2.00 / $12.00$0.10 / $0.20
Context1.1M1M
ReleasedJul 9, 2026Sep 2, 2026
WeightsClosedClosed
API hosts42

Don’t pick one. Route.

Shogo can run GPT-5.6 Terra and Muse Spark 1.3 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo