ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

GLM-4.5vsGLM-5.2

GLM-4.5 and GLM-5.2 do not yet share a public benchmark, so quality cannot be compared apples-to-apples. GLM-4.5 is 2.1x cheaper per token. GLM-5.2 has the larger context window (1M tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
0
no overlap yet
Cheaper per token
GLM-4.5
2.1x cheaper, input + output
Larger context
GLM-5.2
1M tokens
Provider uptime (30d)
Z.ai
text
Context
128K
Max out
98K
Input /1M
$0.6
Output /1M
$2.2
Cached /1M
$0.11
Scores
2 · 2 core
GLM-5.2
Z.ai · released 2026-06-24
text
Context
1M
Max out
131K
Input /1M
$1.4
Output /1M
$4.4
Cached /1M
$0.26
Scores
26 · 9 core
Quality

Benchmark matrix

BenchmarkGLM-4.5GLM-5.2ΔEdge
Reported by both · 0
No public benchmark is reported for both models yet.
Only GLM-4.5 reports · 2
AIME 202593.3% matharenanot reported
LMArena Elo1430.2 not reported
Only GLM-5.2 reports · 25
GPQA Diamondnot reported91.9% epoch
AIME 2026not reported90% matharena
Agents' Last Exam (ALE-CLI)not reported23.8%
AutomationBench v1.0.6not reported26.2%
CyberGymnot reported77.2%
DeepSWE v1.1not reported46.2%
ExploitBenchnot reported24.4%
ExploitGym 2hnot reported29
ExploitGym 6hnot reported39
FrontierMath Tier 4 v2 (Epoch AI run)not reported29.3% epoch
FrontierMath Tiers 1-3 v2 (Epoch AI run)not reported59.2% epoch
FrontierSWEnot reported67.5%
GDPval-AA v2not reported1508
HLE (with tools)not reported54.7%
NL2Reponot reported48.9%
OTIS Mock AIME 2024-2025 (Epoch AI run)not reported86.4% epoch
PostTrainBenchnot reported31.7%
ProgramBench (Almost Solved)not reported9.5%
SimpleQA Verifiednot reported34.2% epoch
SWE-Bench Pronot reported62.1%
SWE-bench Verified (Epoch AI run)not reported78.7% epoch
SWE-Marathon v1.1not reported19.4%
Terminal-Bench 2.1not reported81%
Terminal-Bench 3.0not reported4.6%
Toolathlon Verifiednot reported59.9%
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is GLM-4.5 minus GLM-5.2 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

SpecGLM-4.5GLM-5.2Edge
Context window128K tokens1M tokensGLM-5.2
Max output98K tokens131K tokensGLM-5.2
Input price / 1M$0.6 $1.4 GLM-4.5
Output price / 1M$2.2 $4.4 GLM-4.5
Cached input / 1M$0.11 $0.26 GLM-4.5
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$2.8$5.8GLM-4.5
ModalitiestexttextTie
Released2026-06-24
Cited benchmark scores226
Reliability

Z.ai status

All providers
More matchups

GLM-5.2 vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run GLM-4.5 and GLM-5.2 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.