Migrating from glm-5.1 to glm-5.2

What changes when you move from glm-5.1 to glm-5.2: input price 0%, output price 0%.

From: glm-5.1
Status: Active
Shutdown: none announced
Time left: —

To: glm-5.2
Status: Active
Shutdown: none announced
Time left: —

Prices

USD per 1M tokens, standard tier

Type	glm-5.1	glm-5.2	Change
Input	$1.40first seen October 5, 2026 · source	$1.40first seen October 5, 2026 · source	0%
Cached input	$0.26first seen October 5, 2026 · source	$0.26first seen October 5, 2026 · source	0%
Output	$4.40first seen October 5, 2026 · source	$4.40first seen October 5, 2026 · source	0%

At 1M input and 250K output tokens a month: $2.50 on glm-5.1, $2.50 on glm-5.2 (0%).

Other tiers, prompt-size bands and modalities are on each model's page.

Limits

Tokens

Neither company publishes a context window or output limit for these models.

Benchmarks

Same benchmark and source on both sides; settings named where they differ

Benchmark	glm-5.1	glm-5.2	Change
Arena TextArena, max effort (vs no setting). Ranges don't overlap.	1465 ±4	1475 ±4	+10
Arena Text: CodingArena, max effort (vs no setting).	1512 ±6	1513 ±6	+1 within range
Arena WebDevArena, max effort (vs no setting). Ranges don't overlap.	1508 ±7	1604 ±6	+96
FrontierMath Tiers 1-3Epoch AI, max effort (vs no setting). Other settings not run → 54.7% low effort 24.9% → 42.5% no reasoning	36.8%	59.2%	+22.4 pts
GPQA DiamondEpoch AI, max effort (vs no setting). Other settings not run → 87.9% low effort not run → 71.2% no reasoning	89.9%	91.9%	+2.0 pts
OTIS Mock AIME 2024-2025Epoch AI, max effort (vs no setting). Other settings not run → 75.6% low effort not run → 28.9% no reasoning	93.3%	86.4%	−6.9 pts
SimpleQA VerifiedEpoch AI, max effort (vs no setting).	34.0%	34.2%	+0.2 pts
SWE-bench VerifiedEpoch AI, max effort (vs no setting).	74.2%	78.7%	+4.5 pts

A score from one source is never compared with another source's, and company-reported scores are compared only within the same company; settings named where they differ. Data: Arena (arena.ai), CC BY 4.0; Epoch AI, Capabilities & benchmarking (epoch.ai/benchmarks), CC BY.

Features and parameters

Compare models

Neither company publishes feature facts for these models.

Aliases

All aliases

No aliases point at either model.

Checklist

Change the model string from glm-5.1 to glm-5.2.

Compared from AIWire's published data when this page loaded. Each fact links to the page it came from. Model pages: glm-5.1, glm-5.2.