SShogo Bench

GLM-5.3-Flash vs GPT-5.6 Luna

GPT-5.6 Luna leads GLM-5.3-Flash by 3.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GLM-5.3-Flash wins 5 and GPT-5.6 Luna wins 10 of 15 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGLM-5.3-FlashGPT-5.6 Luna
Agentic coding5459
Long-horizon autonomy5761
Knowledge-work agents4039
Computer use3440
Security & context learning4751

Shared benchmarks

BenchmarkGLM-5.3-FlashGPT-5.6 Luna
APEX-Agentsindex52.8%43.0%
DeepSWEindex63.4%67.2%
CursorBenchindex36.8%35.9%
ARC-AGI91.0%88.0%
ARC-AGI-265.8%59.5%
Chess Puzzles21.0%43.0%
CritPt15.4%20.6%
FrontierCode31.8%39.8%
FrontierMath Tier 417.1%61.0%
FrontierMath Tiers 1-355.8%82.1%
GPQA Diamond90.2%91.6%
OTIS Mock AIME93.9%98.3%
ProofBench21.0%60.0%
SciCode51.6%53.6%
WebDev Arena16071519

Specs and price

GLM-5.3-FlashGPT-5.6 Luna
Agent Index48.3 (#38)51.4 (#34)
Agent $/M$0.16$0.10
Input / output $/M$0.30 / $1.05$0.20 / $1.20
Context1M1.1M
ReleasedSep 18, 2026Jul 9, 2026
WeightsOpenClosed
API hosts425

Don’t pick one. Route.

Shogo can run GLM-5.3-Flash 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