SShogo Bench

Gemini 3.8 Flash vs GLM-5.3-Flash

Gemini 3.8 Flash leads GLM-5.3-Flash by 7.9 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Gemini 3.8 Flash wins 15 and GLM-5.3-Flash wins 1 of 16 shared benchmarks. GLM-5.3-Flash is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGemini 3.8 FlashGLM-5.3-Flash
Agentic coding6354
Long-horizon autonomy6457
Knowledge-work agents4740
Computer use4434
Security & context learning5347

Shared benchmarks

BenchmarkGemini 3.8 FlashGLM-5.3-Flash
APEX-Agentsindex64.3%52.8%
DeepSWEindex73.8%63.4%
FrontierSWEindex19.6%18.1%
CursorBenchindex39.6%36.8%
ARC-AGI98.5%91.0%
ARC-AGI-289.2%65.8%
Chess Puzzles58.0%21.0%
CritPt18.3%15.4%
FrontierCode41.2%31.8%
FrontierMath Tier 422.0%17.1%
FrontierMath Tiers 1-368.4%55.8%
GPQA Diamond95.4%90.2%
OTIS Mock AIME98.9%93.9%
ProofBench48.0%21.0%
SciCode56.6%51.6%
WebDev Arena15681607

Specs and price

Gemini 3.8 FlashGLM-5.3-Flash
Agent Index56.2 (#21)48.3 (#38)
Agent $/M$0.35$0.16
Input / output $/M$0.75 / $3.75$0.30 / $1.05
Context1M1M
ReleasedSep 2, 2026Sep 18, 2026
WeightsClosedOpen
API hosts642

Don’t pick one. Route.

Shogo can run Gemini 3.8 Flash and GLM-5.3-Flash side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo