GLM-5.2

API ID glm-5.2

GLM-5.2 is a Z.ai model, released June 2026. Standard price: $1.40 input and $4.40 output per 1M tokens.

Status: Active
Released: June 2026
Shutdown: none announced
Replacement: —

Z.ai doesn't publish shutdown dates. AIWire checks its model list daily and will flag glm-5.2 if it disappears.

Benchmarks

Scores as each source published them

Comparing with GLM-5.1, the previous model in this line. A score is compared only with the same benchmark from the same source, at the same setting where both have it. Full comparison: prices, limits, features

Arena

Arena	GLM-5.2	GLM-5.1	Change	Rank	Date	Source
Text By category Math 1485 rank 32 vs 1476 (+9, within range) at max effort (vs no setting) Creative writing 1453 rank 34 vs 1447 (+6, within range) at max effort (vs no setting) Instruction following 1469 rank 38 vs 1462 (+7, within range) at max effort (vs no setting) Hard prompts 1496 rank 40 vs 1489 (+7, within range) at max effort (vs no setting) Long prompts 1486 rank 34 vs 1480 (+6, within range) at max effort (vs no setting) Multi-turn 1476 rank 48 vs 1478 (−2, within range) at max effort (vs no setting) Expert 1502 rank 40 vs 1496 (+6, within range) at max effort (vs no setting)	1475 ±4	1465 ±4	+10	34 of 415 49,127 votes	Oct 13, 2026	source
Text: Coding	1513 ±6	1512 ±6	+1 within range	51 of 410 13,235 votes	Oct 13, 2026	source
WebDev	1604 ±6	1508 ±7	+96	28 of 142 16,093 votes	Oct 13, 2026	source
Agent	0.024 ±0.008	not rated		23 of 54 86,843 sessions	Oct 14, 2026	source

Rated as glm-5.2-max (max effort); GLM-5.1 as glm-5.1. Some were rated at different settings; ratings on one arena share one scale, so the difference still reads directly. Text and WebDev ratings come from people voting between two anonymous answers; Text is style-controlled, as on arena.ai. Agent is Arena's separate score for real agent sessions, on its own scale. Data: Arena (arena.ai), CC BY 4.0.

Epoch AI

Benchmark	GLM-5.2	GLM-5.1	Change	Setting	Run	Source
FrontierMath Tier 4	29.3%	not run		max effort	Jun 19, 2026	source
FrontierMath Tiers 1-3 Other settings 54.7% vs not run low effort 42.5% vs 24.9% no reasoning	59.2%	36.8%	+22.4 pts	max effort (vs no setting)	Jun 19, 2026	source
GPQA Diamond Other settings 87.9% vs not run low effort 71.2% vs not run no reasoning	91.9%	89.9%	+2.0 pts	max effort (vs no setting)	Jun 24, 2026	source
OTIS Mock AIME 2024-2025 Other settings 75.6% vs not run low effort 28.9% vs not run no reasoning	86.4%	93.3%	−6.9 pts	max effort (vs no setting)	Jun 25, 2026	source
SimpleQA Verified	34.2%	34.0%	+0.2 pts	max effort (vs no setting)	Aug 27, 2026	source
SWE-bench Verified	78.7%	74.2%	+4.5 pts	max effort (vs no setting)	Jun 25, 2026	source

Run independently by Epoch AI. Data: Epoch AI, Capabilities & benchmarking (epoch.ai/benchmarks), CC BY.

Price history

USD per 1M tokens, standard tier

We started recording this model's prices on Oct 5, 2026. The history grows with each daily check.

Show as a table

From	Type	Price
Oct 5, 2026	Input	$1.40
Oct 5, 2026	Output	$4.40

Prices

USD per 1M tokens, every type and band, by tier

Type	Price $/1M	In effect
Standard
input	$1.40
cached input	$0.26
output	$4.40

Changelog

Changes we detected on Z.ai's pages

No changes recorded yet.

Email alerts

Get an email if glm-5.2 changes

We'll email you the same morning we spot a deprecation notice, a new or moved shutdown date, a price change, or the model leaving Z.ai's model list.

No password: one email to confirm, and you can unsubscribe any time. Privacy

Last checked Oct 15, 2026, 2:00 am ET · Sources: Z.ai models, Z.ai pricing