Inference
EN

Claude Haiku 5.5 vs GLM-5.3

Change models
Claude Haiku 5.5Anthropic
GLM-5.3Z.AI
IndexAll categories, out of 100.
66.3 ±2.8 best
65.8 ±3.5
GapIndex points below the top.
Top
−0.5 · no clear gap
CoverageIndex weight that has results.
95% best
85%
SpeedOutput tokens per second.
159/s best
86/s
EloLMArena rating and rank.
1399
1476 best
ReleasedMonth of release.
2026-10
2026-08
Input / 1MUS dollars per 1M input tokens.
$0.10
$0.07 best
Output / 1MUS dollars per 1M output tokens.
$0.50 best
$1.40
Batch output / 1MBatch price per 1M output tokens.
$0.25 best
$0.989
1K chats2K tokens in, 500 out, each.
$0.45 best
$0.84
1K documents20K tokens in, 1K out, each.
$2.50 best
$2.80
1K agent tasks200K tokens in, 20K out, each.
$30.00 best
$41.98
ContextMaximum tokens in one request.
1M
1.05M best
AgenticMulti-step tasks with tools.
68.6 best
68.2
CodingCode writing and repair.
71.0 best
65.7
ReasoningLogic problems and puzzles.
67.8 best
64.5
MultimodalTasks with images and text.
51.9
—
KnowledgeFacts and expert knowledge.
54.5
59.3 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
37.2
66.8 best
MathMath problems.
68.3 best
63.1

Benchmarks

57 shared
Only Claude Haiku 5.5: 27Only GLM-5.3: 39
Claude Haiku 5.5
GLM-5.3
Artificial Analysis Humanity's Last ExamKnowledge
44.4% best28 / 222
42.3%39 / 222
Artificial Analysis Long Context ReasoningReasoning
82.7% best23 / 217
79.7%45 / 217
GPQA diamond · Epoch AIKnowledge
89.6%44 / 217
90.9% best35 / 217
Critical Physics TasksReasoning
18.9%33 / 214
19.1% best32 / 214
OTIS Mock AIME 2024-2025Math
98.9% best22 / 200
91.1%58 / 200
Artificial Analysis Omniscience AccuracyKnowledge
34.7% best68 / 164
32.3%73 / 164
GDPval-AA normalizedAgentic
56.0%15 / 133
57.5% best12 / 133
Vibe Code Bench v1.1Coding
90.4% best4 / 123
78.1%36 / 123
Artificial Analysis SciCodeCoding
55.0%29 / 114
59.0% best12 / 114
MedScribeHealthcare
89.5% best6 / 109
88.8%11 / 109
MedCodeHealthcare
51.5% best15 / 106
42.9%49 / 106
SimpleQA VerifiedKnowledge
23.8%75 / 91
41.0% best51 / 91
Chess PuzzlesReasoning
55.0% best16 / 90
37.0%39 / 90
FrontierMath-Tiers-1-3-v2-PrivateMath
75.1% best23 / 90
68.8%31 / 90
Legal Research BenchLegal
43.8%25 / 90
49.0% best10 / 90
Tax Agent BenchFinance
31.9%27 / 82
43.3% best7 / 82
LiveBench Data AnalysisReasoning
52.4%79 / 80
70.2% best56 / 80
LiveBench ReasoningReasoning
83.5%45 / 80
85.8% best35 / 80
Mystery Game PuzzlesReasoning
30.0%30 / 80
33.0% best28 / 80
Finance Agent (v2)Finance
54.1%28 / 79
55.8% best22 / 79