SShogo Bench

GLM-5.3 vs GPT-5.6 Luna

GLM-5.3 leads GPT-5.6 Luna by 3.8 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GLM-5.3 wins 9 and GPT-5.6 Luna wins 8 of 18 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGLM-5.3GPT-5.6 Luna
Agentic coding6259
Long-horizon autonomy6861
Knowledge-work agents4339
Computer use4040
Security & context learning5151

Shared benchmarks

BenchmarkGLM-5.3GPT-5.6 Luna
APEX-Agentsindex56.6%43.0%
DeepSWEindex69.0%67.2%
Vending-Bench 2index$8,164$4,095
CursorBenchindex42.6%35.9%
Chess Puzzles37.0%43.0%
CritPt19.1%20.6%
DTBench87.7%89.1%
FrontierCode40.1%39.8%
FrontierMath Tier 429.3%61.0%
FrontierMath Tiers 1-368.8%82.1%
GPQA Diamond90.9%91.6%
LMCA55.5%48.5%
OTIS Mock AIME91.1%98.3%
ProofBench49.0%60.0%
SciCode59.0%53.6%
SimpleBench66.2%46.8%
SimpleQA Verified41.0%41.0%
WebDev Arena16141519

Specs and price

GLM-5.3GPT-5.6 Luna
Agent Index55.2 (#25)51.4 (#34)
Agent $/M$0.61$0.10
Input / output $/M$1.20 / $4.18$0.20 / $1.20
Context1M1.1M
ReleasedAug 14, 2026Jul 9, 2026
WeightsOpenClosed
API hosts495

Don’t pick one. Route.

Shogo can run GLM-5.3 and GPT-5.6 Luna side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo