Frontier BenchAI model benchmarks · Jul 2026
ModelsCompareLabsMethodology
All models
ZA

GLM-4.6

Zhipu AI (Z.ai)

Cost-effective coding & agentic model. Strong math for the price.

MidOpen weights200K contextReleased Sep 2025textvisioncode

API price · per 1M tokens

$0.43/$1.74

input / output

≈ $0.76/M blended

Benchmarks

Bar shows this model's score; tick shows the best score across all tracked models for context.

SWE-bench VerifiedReal GitHub issues, human-verified patches. The standard coding-agent benchmark.
68.0/ best 90.1
GPQA DiamondGoogle-proof graduate-level science Q&A. General reasoning proxy.
81.0/ best 94.4
MMLU-ProMulti-task language understanding across 120+ academic subjects.
82.9/ best 93.0
HumanEval+Functional code generation from docstrings.
84.0/ best 96.0
MATH-500Competition-style math problems.
88.0/ best 97.8
AIME 2025American Invitational Mathematics Examination.
93.9/ best 96.5

Built by GLM-5.2 · Scores are directional, drawn from public mid-2026 benchmarks.

SWE-bench, GPQA, MMLU, AIME, HLE etc. are trademarks of their respective owners.