SShogo Bench

Claude Fable 5.2 vs GLM-5.3

Claude Fable 5.2 leads GLM-5.3 by 21.2 points on the Shogo Agent Index. Claude Fable 5.2 wins 17 and GLM-5.3 wins 0 of 17 shared benchmarks. GLM-5.3 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarClaude Fable 5.2GLM-5.3
Agentic coding8662
Long-horizon autonomy7868
Knowledge-work agents6743
Computer use7740
Security & context learning6551

Shared benchmarks

BenchmarkClaude Fable 5.2GLM-5.3
APEX-Agentsindex80.1%56.6%
Vending-Bench 2index$9,628$8,164
FrontierSWEindex69.9%30.2%
CursorBenchindex60.0%42.6%
PostTrainBenchindex44.6%34.5%
Chess Puzzles79.0%37.0%
CritPt35.7%19.1%
DTBench99.2%87.7%
FrontierMath Tier 4100.0%29.3%
FrontierMath Tiers 1-395.1%68.8%
GPQA Diamond96.0%90.9%
LMCA70.6%55.5%
OTIS Mock AIME99.7%91.1%
ProofBench100.0%49.0%
SciCode67.7%59.0%
SimpleBench94.9%66.2%
SimpleQA Verified94.0%41.0%

Specs and price

Claude Fable 5.2GLM-5.3
Agent Index76.4 (#1)55.2 (#25)
Agent $/M$3.21$0.61
Input / output $/M$8.00 / $40.00$1.20 / $4.18
Context1M1M
ReleasedNov 9, 2026Aug 14, 2026
WeightsClosedOpen
API hosts449

Don’t pick one. Route.

Shogo can run Claude Fable 5.2 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