SShogo Bench

Claude Sonnet 5 vs GLM-5.3

GLM-5.3 leads Claude Sonnet 5 by 2.8 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Claude Sonnet 5 wins 5 and GLM-5.3 wins 12 of 18 shared benchmarks. GLM-5.3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Sonnet 5GLM-5.3
Agentic coding5762
Long-horizon autonomy6568
Knowledge-work agents4343
Computer use4040
Security & context learning5151

Shared benchmarks

BenchmarkClaude Sonnet 5GLM-5.3
APEX-Agentsindex54.5%56.6%
DeepSWEindex53.8%69.0%
Vending-Bench 2index$6,378$8,164
CursorBenchindex34.1%42.6%
Chess Puzzles45.0%37.0%
CritPt16.9%19.1%
DTBench92.5%87.7%
FrontierCode42.7%40.1%
FrontierMath Tier 429.3%29.3%
FrontierMath Tiers 1-365.6%68.8%
GPQA Diamond90.5%90.9%
LMCA50.0%55.5%
OTIS Mock AIME94.7%91.1%
ProofBench77.0%49.0%
SciCode54.3%59.0%
SimpleBench60.6%66.2%
SimpleQA Verified33.7%41.0%
WebDev Arena15371614

Specs and price

Claude Sonnet 5GLM-5.3
Agent Index52.4 (#32)55.2 (#25)
Agent $/M$0.92$0.61
Input / output $/M$2.00 / $10.00$1.20 / $4.18
Context1M1M
ReleasedJun 30, 2026Aug 14, 2026
WeightsClosedOpen
API hosts549

Don’t pick one. Route.

Shogo can run Claude Sonnet 5 and GLM-5.3 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo