GLM 5

ModelOpen sourceActive

by Z.ai (Zhipu AI) · family “glm”

GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows. Built for expert developers, it delivers production-grade performance on large-scale programming tasks, rivaling leading... (description from the OpenRouter listing)

in: textout: textReasoningTool useStructured output

Epoch Capabilities Index

145.8 #74 of 268

90% CI 143.7 – 147.7

Listing ↗

Released

11 Feb 2026

Context window

205K tokens

Max output

128K tokens

Input $ / 1M tokens

$0.60

Output $ / 1M tokens

$1.92

Cached input $ / 1M

$0.12

Each benchmark shown separately with its own source

Benchmark results (12)

All benchmarks

Benchmark	Domain	Score	vs best recorded	Setting	Run	Source
CL-bench	long-context	18.7%	67%	—	—	External ↗
Chess Puzzles	games	10.0% ±3.0	14%	—	12 Feb 2026	Eval log ↗
FrontierMath-Tier-4-2025-07-01-Privatesuperseded	math	2.1% ±2.1	4%	—	19 Feb 2026	Epoch ↗
Terminal Bench	agents	52.4% ±2.6	62%	agent: Terminus 2	11 Feb 2026	External ↗
ARC-AGI-2	reasoning	4.9%	5%	—	—	External ↗
FrontierMath-2025-02-28-Privatesuperseded	math	16.4% ±2.2	31%	—	19 Feb 2026	Epoch ↗
WeirdML	coding	48.2% ±0.0	51%	—	—	External ↗
OTIS Mock AIME 2024-2025	math	80.0% ±6.0	80%	—	12 Feb 2026	Eval log ↗
SimpleBench	reasoning	53.2%	65%	—	—	External ↗
SWE-Bench verified	coding	72.1% ±2.1	86%	—	15 Feb 2026	Eval log ↗
GPQA diamond	science	87.8% ±2.3	92%	—	12 Feb 2026	Eval log ↗
ARC-AGI	reasoning	44.7%	45%	—	—	External ↗

Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.

API list price over time

Price history

$ per 1M tokens

8 observations (OpenRouter listing + Internet Archive snapshots). History accumulates with every ingest run; a single point means no change has been observed yet.