Compare Models
GLM 5.3 Flash

Release Date: Aug 26, 2026

Developer zAI 🇨🇳
Context Window 1M
Max Output Tokens 128k
Token Costs (in/out) $0.07/0.25
Weights Open
Input Modalities

Accuracy

47.22 % ± 1.45

Cost / Test (Vals Index)

$ 0.993

Latency

71 min 7 s

Vals Index
BenchmarksAccuracyRankings
Vals Index

0.0%

±1.45
31/57
Code Migration

0.0%

±4.21
34/59
CyberBench

0.0%

±5.93
12/25
EMB

0.0%

±3.56
31/57
Finance Agent (v2)

0.0%

±2.06
9/60
Legal Research Bench

0.0%

±3.46
9/60
MedScribe

0.0%

±1.91
4/94
ProofBench v1.1

0.0%

±4.09
23/31
TaxEval v2

0.0%

±0.84
16/145
Vibe Code Bench v1.1

0.0%

±5.28
55/95
Harvey's Legal Agent Benchmark

0.0%

±1.65
20/61
GPQA Diamond

0.0%

±2.13
43/138
IOI

0.0%

±2.62
13/27
LiveCodeBench

0.0%

±1.05
69/143
LegalBench

0.0%

±0.43
41/144
MMLU Pro

0.0%

±0.34
49/138
MMMU Pro

0.0%

±0.83
23/93
SkillsBench

0.0%

±4.43
30/35
SWE-bench

0.0%

±1.21
10/88
Terminal-Bench 2.1

0.0%

±2.25
31/65
Terminal-Bench Science

0.0%

±0.00
25/25
Proprietary BenchmarksAcademic BenchmarksIndustry Partners
Vals
Default Provider : Fireworks AI
Temperature: 1
Top P: 0.95
Top K: Default
Max Output Tokens: 128,000
Reasoning Effort: max