Zhipu AI · mit

GLM-4.5

GLM-4.5 by Zhipu AI appears in 2 sources with Reasoning at 33.95. Best read for Code quality, Coding, High intelligence.

CreatorZhipu AI

Release date2025-07-28

Knowledge cutoffNot published

ContextUnknown

Input priceUnknown

Output priceUnknown

Modalitytext

CountryCN

Metrics

All source-backed metrics

Open in leaderboard

LLM Stats Rank74ranking · source_rank

Reasoning33.95reasoning · index_reasoning

Math34.25math · index_math

Coding20.72coding · index_code

Research-4.57research · index_search

Writing25.15writing · index_communication

Vision8.19multimodal · index_vision

Tool calling25.62tool_calling · index_tool_calling

Finance41.64domain · index_finance

Legal37.52domain · index_legal

Healthcare37.3domain · index_healthcare

GPQA79.1 %reasoning · gpqa_score

SWE-bench Verified64.2 %coding · swe_bench_verified_score

Code Arena744.07 %coding · coding_arena_score

Humanity Last Exam14.4 %reasoning · hle_score

BrowseComp26.4 %research · browsecomp_score

Terminal Bench37.5 %coding · terminal_bench_score

TAU2 Retail79.7 %agent · tau_bench_retail_score

SciCode41.7 %coding · scicode_score

Parameters355,000,000,000 paramsmodel · params

Arena Rating1,428.95metric

Arena Rank52metric

Vote Count24,340metric

Evidence

Citations and source overlap

Arena AILM Arena text leaderboard (overall) mirrored in the public leaderboard dataset.Retrieved 2026-05-20 LLM StatsPublic full LLM leaderboard advanced explorer snapshot, including ranking, benchmark indexes, pricing, speed, context, license, and model metadata.Retrieved 2026-06-01

FAQ

How should I read this profile?

Treat this as a source-backed model dossier, not an EvalKit-run verification. The public values are replicated from linked sources and kept source-scoped.

Is GLM-4.5 verified by EvalKit?

No. EvalKit currently shows 0 verified rows until real run evidence exists.

Why can metrics disagree?

Different sources test different tasks, dates, prompts, and aggregation methods. EvalKit keeps those differences visible instead of merging them into a fake universal score.