Inference
EN

GLM-5.3 vs GLM-5.3-Flash

Change models
GLM-5.3Z.AI
GLM-5.3-FlashZ.AIBudget
IndexAll categories, out of 100.
65.8 ±3.5 best
63.1 ±2.2
GapIndex points below the top.
Top
−2.7 · no clear gap
CoverageIndex weight that has results.
85%
100% best
SpeedOutput tokens per second.
86/s
122/s best
EloLMArena rating and rank.
1476 best
1471
ReleasedMonth of release.
2026-08
2026-08
Input / 1MUS dollars per 1M input tokens.
$0.07
$0.04 best
Output / 1MUS dollars per 1M output tokens.
$1.40
$0.078 best
Batch output / 1MBatch price per 1M output tokens.
$0.989
—
1K chats2K tokens in, 500 out, each.
$0.84
$0.119 best
1K documents20K tokens in, 1K out, each.
$2.80
$0.876 best
1K agent tasks200K tokens in, 20K out, each.
$41.98
$9.55 best
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
68.2
69.7 best
CodingCode writing and repair.
65.7 best
62.3
ReasoningLogic problems and puzzles.
64.5 best
58.9
MultimodalTasks with images and text.
—
65.0
KnowledgeFacts and expert knowledge.
59.3 best
57.7
MultilingualTasks in many languages.
—
64.0
InstructionTasks with strict prompt rules.
66.8 best
33.5
MathMath problems.
63.1 best
57.3

Benchmarks

67 shared
Only GLM-5.3: 29Only GLM-5.3-Flash: 23
GLM-5.3
GLM-5.3-Flash
GPQA diamond · Epoch AIKnowledge
90.9% best35 / 217
90.2%40 / 217
OTIS Mock AIME 2024-2025Math
91.1%58 / 200
93.9% best49 / 200
LegalBenchLegal
84.8% best29 / 149
83.9%45 / 149
TaxEval v2Finance
72.4%62 / 143
75.6% best14 / 143
LiveCodeBenchCoding
80.5%67 / 142
80.5%67 / 142
GPQA Diamond · Vals AIKnowledge
88.1% best35 / 135
86.4%43 / 135
MMLU ProKnowledge
86.8% best38 / 135
86.1%49 / 135
Vibe Code Bench v1.1Coding
78.1% best36 / 123
30.8%82 / 123
MedScribeHealthcare
88.8%11 / 109
88.9% best9 / 109
Chess PuzzlesReasoning
37.0% best39 / 90
21.0%59 / 90
FrontierMath-Tiers-1-3-v2-PrivateMath
68.8% best31 / 90
55.8%49 / 90
Legal Research BenchLegal
49.0% best10 / 90
45.2%22 / 90
SWE-benchCoding
95.4% best5 / 87
92.0%10 / 87
Terminal-Bench 2.1 · BenchLMAgentic
88.2% best10 / 86
84.3%22 / 86
Tax Agent BenchFinance
43.3% best7 / 82
31.6%28 / 82
LiveBench Data AnalysisReasoning
70.2%56 / 80
76.4% best35 / 80
LiveBench ReasoningReasoning
85.8% best35 / 80
77.6%65 / 80
Mystery Game PuzzlesReasoning
33.0% best28 / 80
8.0%66 / 80
Finance Agent (v2)Finance
55.8%22 / 79
57.9% best13 / 79
Code MigrationCoding
44.2% best24 / 77
20.5%52 / 77