Z.aiopen weightsCurrentReasoning

GLM-5.3 Flash

Name: GLM-5.3 Flash
Author: Z.ai

Released

Aug 26, 2026

Price in / out

—

Blended $/1M

—

Context

Parameters: 320B / 18B active. License: MIT.

Benchmark scoresvs GLM-5.3

Benchmark	Score	Δ vs prior	Rank (current)	Configuration	Source
Terminal-Bench 2.1	62.9%	-8.6	#17 of 22	leaderboard	benchlm.ai ↗
GPQA Diamond	86.4%	-1.7	#16 of 21	Vals AI run, max reasoning · independent run	vals.ai ↗
Humanity's Last Exam	39.9%	-2.4	#18 of 31	Artificial Analysis run, no tools (2,158 text-only questions); Z.ai reports 55.3 with tools · independent run	artificialanalysis.ai ↗
MMLU-Pro	86.1%	-0.7	#8 of 12	Vals AI run, max reasoning · independent run	vals.ai ↗

Version

$ in / out

Terminal-Bench

SWE-bench

GPQA

HLE

AIME

τ²-bench

GLM-4.6Superseded

—

76.9%

GLM-4.7Superseded

—

73.8%

—

24.8%

—

95.9%

GLM-5Superseded

—

77.8%

86%

50.4%

95.8%

98.2%

GLM-5.1Superseded

—

56.9%

—

86.2%

52.3%

95.3%

97.7%

GLM-5.2Superseded

$1.4 / $4.4

67.8%

—

91.2%

54.7%

99.2%

99.1%

GLM-5.3

$1.4 / $4.4

71.5%

—

88.1%

42.3%

—

GLM-5.3 Flash

—

62.9%

—

86.4%

39.9%

—

GLM-5.3 Flash

Benchmark scoresvs GLM-5.3

GLM versions

Compare

Shipping with it?

Current open-weight peers

Version	Released	$ in / out	Terminal-Bench	SWE-bench	GPQA	HLE	AIME	τ²-bench
GLM-4.6Superseded	2025-09-30	—	—	—	—	—	—	76.9%
GLM-4.7Superseded	2025-12-22	—	—	73.8%	—	24.8%	—	95.9%
GLM-5Superseded	2026-02-11	—	—	77.8%	86%	50.4%	95.8%	98.2%
GLM-5.1Superseded	2026-04-15	—	56.9%	—	86.2%	52.3%	95.3%	97.7%
GLM-5.2Superseded	2026-06-17	$1.4 / $4.4	67.8%	—	91.2%	54.7%	99.2%	99.1%
GLM-5.3	2026-08-14	$1.4 / $4.4	71.5%	—	88.1%	42.3%	—	—
GLM-5.3 Flash	2026-08-26	—	62.9%	—	86.4%	39.9%	—	—