Inference
EN

GLM-5.3 vs GPT-6 Luna

Change models
GLM-5.3Z.AI
GPT-6 LunaOpenAI
IndexAll categories, out of 100.
65.8 ±3.5 best
65.5 ±2.1
GapIndex points below the top.
Top
−0.3 · no clear gap
CoverageIndex weight that has results.
85%
100% best
SpeedOutput tokens per second.
86/s best
66/s
EloLMArena rating and rank.
1476 best
1388
ReleasedMonth of release.
2026-08
2026-09
Input / 1MUS dollars per 1M input tokens.
$0.07 best
$0.10
Output / 1MUS dollars per 1M output tokens.
$1.40
$0.50 best
Batch output / 1MBatch price per 1M output tokens.
$0.989
$0.25 best
1K chats2K tokens in, 500 out, each.
$0.84
$0.45 best
1K documents20K tokens in, 1K out, each.
$2.80
$2.50 best
1K agent tasks200K tokens in, 20K out, each.
$41.98
$30.00 best
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
68.2 best
66.9
CodingCode writing and repair.
65.7
66.8 best
ReasoningLogic problems and puzzles.
64.5 best
64.3
MultimodalTasks with images and text.
—
65.0
KnowledgeFacts and expert knowledge.
59.3
63.5 best
MultilingualTasks in many languages.
—
53.3
InstructionTasks with strict prompt rules.
66.8 best
39.8
MathMath problems.
63.1
68.5 best

Benchmarks

63 shared
Only GLM-5.3: 33Only GPT-6 Luna: 38
GLM-5.3
GPT-6 Luna
Artificial Analysis Humanity's Last ExamKnowledge
42.3% best39 / 222
38.5%54 / 222
Artificial Analysis Long Context ReasoningReasoning
79.7%45 / 217
83.3% best13 / 217
GPQA diamond · Epoch AIKnowledge
90.9% best35 / 217
90.5%39 / 217
Critical Physics TasksReasoning
19.1%32 / 214
19.4% best31 / 214
OTIS Mock AIME 2024-2025Math
91.1%58 / 200
98.9% best22 / 200
Artificial Analysis Omniscience AccuracyKnowledge
32.3%73 / 164
39.6% best54 / 164
GDPval-AA normalizedAgentic
57.5% best12 / 133
46.6%36 / 133
Vibe Code Bench v1.1Coding
78.1%36 / 123
81.6% best26 / 123
Artificial Analysis SciCodeCoding
59.0% best12 / 114
54.6%32 / 114
MedScribeHealthcare
88.8% best11 / 109
83.7%42 / 109
MedCodeHealthcare
42.9%49 / 106
44.7% best40 / 106
SimpleQA VerifiedKnowledge
41.0%51 / 91
41.4% best49 / 91
Chess PuzzlesReasoning
37.0% best39 / 90
36.0%42 / 90
FrontierMath-Tiers-1-3-v2-PrivateMath
68.8%31 / 90
78.9% best21 / 90
Legal Research BenchLegal
49.0% best10 / 90
30.3%54 / 90
Tax Agent BenchFinance
43.3% best7 / 82
19.9%55 / 82
LiveBench Data AnalysisReasoning
70.2%56 / 80
73.4% best45 / 80
LiveBench ReasoningReasoning
85.8% best35 / 80
81.8%54 / 80
Mystery Game PuzzlesReasoning
33.0% best28 / 80
7.0%69 / 80
Finance Agent (v2)Finance
55.8% best22 / 79
49.9%45 / 79