SShogo Bench

GLM-4.7

Mature GLM model for dependable coding, reasoning, and structured agent tasks

Shogo Agent Index

Estimated#71 of 92
30.0range 21.3–38.8 · 19% of weight measured
Agentic coding
30
23% measured
Long-horizon autonomy
47
30% measured
Knowledge-work agents
23
estimated
Computer use
16
estimated
Security & context learning
29
50% measured

Price

Input
$0.45/M
Output
$2.09/M
Cache read
$0.090/M
3:1 blend
$0.86/M
Agent-effective
$0.24/M

Cheapest host for agents: DeepInfra at $0.16/M.

Agent-effective = 20 input tokens per output token, 85% cache hits. Why.

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 63
Terminal-Bench 2.033.4%8%Terminus 2
DeepSWEest.6%predicted 21
FrontierSWEest.5%predicted 3
CursorBenchest.4%predicted 24
GSO-Benchest.4%predicted 6
METR Time Horizonsest.10%predicted 62
Vending-Bench 2$2,3776%source
PostTrainBenchest.4%predicted 16
APEX-Agentsest.8%predicted 21
GDPvalest.7%predicted 27
Remote Labor Indexest.5%predicted 1
DeepResearch Benchest.5%predicted 44
OSWorld 2.0est.6%predicted 2
OSWorldest.4%predicted 38
Cybenchest.5%predicted 41
CL-bench15.9%5%source

Other benchmarks

BenchmarkCategoryScoreSource
Chess PuzzlesReasoning21.0%source
CritPtReasoning1.7%source
GPQA DiamondReasoning83.3%source
OTIS Mock AIMEMath83.3%source
ProofBenchMath6.0%source
SciCodeCoding45.1%source
SimpleBenchReasoning47.7%source
SimpleQA VerifiedKnowledge32.2%source
WebDev ArenaCoding1435source

API hosts

Live prices and uptime for every host serving GLM-4.7 via OpenRouter.

HostInputOutputCache readAgent $/MUptime 24hQuant
DeepInfra$0.32$1.40$0.064$0.16—fp4
DeepInfra$0.40$1.75$0.080$0.21100.0%fp4
Venice$0.40$1.93$0.080$0.21100.0%fp4
Novita$0.60$2.20$0.11$0.2899.4%fp8
Z.AI$0.60$2.20$0.11$0.28100.0%fp4
Mancer 2$0.70$2.50$0.35$0.5099.5%fp4