SShogo Bench

GLM-5.3-Flash vs GPT-6 Astra

GPT-6 Astra leads GLM-5.3-Flash by 22.6 points on the Shogo Agent Index. GLM-5.3-Flash wins 0 and GPT-6 Astra wins 16 of 16 shared benchmarks. GLM-5.3-Flash is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGLM-5.3-FlashGPT-6 Astra
Agentic coding5480
Long-horizon autonomy5780
Knowledge-work agents4056
Computer use3467
Security & context learning4763

Shared benchmarks

BenchmarkGLM-5.3-FlashGPT-6 Astra
APEX-Agentsindex52.8%64.7%
DeepSWEindex63.4%74.1%
FrontierSWEindex18.1%65.5%
PostTrainBenchindex36.8%41.9%
ARC-AGI91.0%98.5%
ARC-AGI-265.8%95.0%
Chess Puzzles21.0%76.0%
CritPt15.4%31.7%
FrontierCode31.8%53.3%
FrontierMath Tier 417.1%97.6%
FrontierMath Tiers 1-355.8%93.7%
GPQA Diamond90.2%95.8%
OTIS Mock AIME93.9%100.0%
ProofBench21.0%99.0%
SciCode51.6%56.5%
WebDev Arena16071800

Specs and price

GLM-5.3-FlashGPT-6 Astra
Agent Index48.3 (#38)70.9 (#4)
Agent $/M$0.16$4.62
Input / output $/M$0.30 / $1.05$10.00 / $50.00
Context1M1.1M
ReleasedSep 18, 2026Sep 4, 2026
WeightsOpenClosed
API hosts426

Don’t pick one. Route.

Shogo can run GLM-5.3-Flash and GPT-6 Astra side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo