LLMcompare

Search

Search for a command to run...

A · open · GLM

GLM-4.7

Zhipu AI

B · open · GLM

GLM-5.2

Zhipu AI

1:13

overall capability wins (0 ties across 14 scored). GLM-5.2 leads overall.

  • Reasoning1:5
  • Coding0:4
  • Arena0:1
  • Tool use0:1
  • Pricing3:0
  • Speed1:1
  • Specs0:2

Verdict

GLM-5.2 leads coding (widest gap: +157.1 on WebDev Elo); GLM-5.2 ranks higher on LMArena (+30 Elo); GLM-5.2 offers ~1.6x larger context (200K vs 128K).

  • GLM-5.2 leads coding (widest gap: +157.1 on WebDev Elo)
  • GLM-5.2 ranks higher on LMArena (+30 Elo)
  • GLM-5.2 offers ~1.6x larger context (200K vs 128K)
  • GLM-5.2 allows ~2.0x more max output (33K vs 16K)
  • GLM-4.7 is ~6.1x cheaper on a blended token basis than GLM-5.2

Cost for 1M in + 250K out

Illustrative chat workload at primary-provider list prices.

GLM-4.7
$0.40
GLM-5.2
$2.50

GLM-4.7 is about 6.3x cheaper on this workload.

Price per 1M tokens

Full comparison

Organization

Zhipu AI

Zhipu AI

Family

GLM

GLM

License

MIT

MIT

Open weights

Yes

Yes

Release

Dec 1, 2025

Jun 1, 2026

Knowledge cutoff

-

-

API / provider

Zhipu / Z.ai

Zhipu / Z.ai

Modalities

text → text

text, image → text

Specs

Context window

128K

200K

B +72K

Max output

16K

33K

B +16K

Parameters

—

—

—

Pricing

Input $/1M

$0.20

$1.40

A +$1.20

Output $/1M

$0.80

$4.40

A +$3.60

Blended $/1M (3∶1)

$0.35

$2.15

A +$1.80

Speed

tok/s

90

168

B +78

TTFT (s)

0.3

0.35

A +0.05

Reasoning

MMLU-Pro

—

—

—

GPQA Diamond

85.9

89.5

B +3.6 pts

Humanity's Last Exam

27.4

41.1

B +13.7 pts

AIME 2025

—

—

—

MATH-500

—

—

—

Humanity's Last Exam (with tools)

—

—

—

AA-Omniscience Accuracy

29.3

24.3

A +5.0 pts

AA-LCR v1.1

71.0

78.3

B +7.3 pts

CritPt

1.7

20.9

B +19.2 pts

MMMU-Pro

—

—

—

IFBench

67.9

73.3

B +5.4 pts

Chartography

—

—

—

Chartography (With Tools)

—

—

—

Coding

SWE-bench Verified

—

—

—

SWE-bench Pro

—

—

—

SWE-bench Multilingual

—

—

—

LiveCodeBench

—

—

—

Terminal-Bench 2.1

45.3

77.9

B +32.6 pts

Aider Polyglot

—

—

—

Terminal-Bench 3

—

4.6

—

BigCodeBench

—

—

—

SciCode

45.1

51.2

B +6.1 pts

CursorBench

—

55.0

—

SWE-Rebench

45.5

57.0

B +11.5 pts

NL2Repo-Bench

—

—

—

DeepSWE

—

44.0

—

WebDev Arena

1,435

1,592

B +157 Elo

Terminal-Bench 4.0

—

—

—

CursorBench 4.0

—

—

—

FrontierCode v1.1 (Main)

—

—

—

Arena

LMArena Elo

1,442

1,472

B +30 Elo

GLM-4.7

Context
128K / 16K out
Parameters
—
Price
$0.20 / $0.80
Speed
90 tok/s · 0.3s TTFT
Modalities
text → text
License
MIT

Prior GLM-4.x open checkpoint still useful for cost-sensitive bilingual apps.

GLM-5.2

Context
200K / 33K out
Parameters
—
Price
$1.40 / $4.40
Speed
168 tok/s · 0.35s TTFT
Modalities
text, image → text
License
MIT

Open MIT GLM-5.2 - competitive open Arena Elo and strong Chinese/English bilingual performance.

Benchmark charts

Winner bars are emphasized. Per-benchmark deltas sit above each chart.

Reasoning & knowledge

GPQA: B +3.6HLE: B +13.7CritPt: B +19.2AA-LCR: B +7.3Omniscience Acc.: A +5.0IFBench: B +5.4

Coding

TermBench: B +32.6SciCode: B +6.1SWE-Rebench: B +11.5WebDev Elo: B +157

Tool use & function calling

τ³-Banking: B +22.4

Arena

Arena Elo: B +30

Pick a different pair · Back to catalog