Frontier Log

Z.aiModel

GLM-4.6

Made by
Z.ai
Released
Announced in
GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilitiesz.ai
Event
GLM-4.6
API record
/api/v1/models/glm-4-6

Benchmark results

BenchmarkScoreSource check
Humanity's Last ExamNo tools · 128K context17.2%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
Humanity's Last ExamWith tools · 128K context30.4%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
SWE-bench Verified128K context68.0%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 4 other lab documents, the first 3 linked: 1 2 3
GPQA Diamondno tools; 128K context81.0%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
BrowseComp128K context45.1%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 3 other lab documents: 1 2 3
with context management · reported later, in GLM-4.7: Advancing the Coding Capability57.5%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
Terminal-Bench 2.0reported later, in GLM-4.7: Advancing the Coding Capability24.5%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
AIME 2025No tools · 128K context93.9%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
AIME 2025With tools · 128K context98.6%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
Terminal-Bench 1.0128K context40.5%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
MMLU-Proreported later, in GLM-4.7: Advancing the Coding Capability83.2%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
HMMT February 2025No tools · reported later, in GLM-4.7: Advancing the Coding Capability89.2%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
LiveCodeBenchLiveCodeBench v6; no tools; 128K context82.8%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
Sources

Reported by Z.ai

Events

All models