메인 콘텐츠로 건너뛰기

Claude Haiku 5.5 (Max)

AnthropicClaudeProprietary

설명

Claude Haiku 5.5 is Anthropic's cheapest, fastest, and most capable small model to date, and the successor to Claude Haiku 4.5. It targets high-volume, cost-sensitive work (summaries, compaction, classification, database queries) and pairs with Opus 5.5 / Sonnet 5.5 as a coding subagent; it is especially strong for speed-sensitive tasks like live support and browser use. It is the first Haiku-class model with adjustable effort (low/medium/high/xhigh/max; Claude Platform default medium). It accepts text and image input (including file/PDF) with text output, supports multilingual and vision workloads, tool use, and adaptive thinking. Context window is 1M tokens with 128K max output; knowledge cutoff is June 2026. Claude Platform pricing is tiered by prompt length: prompts up to 100K tokens are $0.10/$0.50 per million input/output (cache read $0.01, 5m cache write $0.125); prompts over 100K are $0.50/$2.50 (cache read $0.05, 5m cache write $0.625). Catalog provider pricing stores the <=100K tier. Capability scores are self-reported from the Claude Haiku 5.5 System Card (Table 8.1.A and §8) and the launch announcement. Available on the Claude API as `claude-haiku-5-5`.

출시일
2026-10-07
파라미터
—
컨텍스트 길이
1.0M
모달리티
image, pdf, text

능력 레이더

44
general
55
coding
50
reasoning
51
science
30
agents
60
multimodal

랭킹

도메인#순위점수소스
에이전트형 역량42
50.0
LS
코딩 랭킹15
93.0
AA
종합 랭킹23
74.0
AA
수학 추론39
86.0
LB
멀티모달 랭킹83
50.0
LS
추론29
85.0
LB
과학49
76.0
AA

벤치마크 점수 (LLM Stats)

(LLM Stats (zeroeval))

Agents

AA-Briefcase v1.11578.00 / 3000자체 보고
Program Bench82.0%자체 보고
OfficeQA73.5%자체 보고
OfficeQA Pro60.3%자체 보고

Biology

LatchBio SpatialBench Verified67.7%자체 보고
Protocols Understanding (V2)64.1%자체 보고
Protocols Troubleshooting58.0%자체 보고
Medicinal Chemistry (ADME)58.0%자체 보고
Protein Design — Library Ranking45.3%자체 보고
Biomedical image analysis43.5%자체 보고
Protein Design — Sequence Generation33.1%자체 보고
Morphology-to-molecule matching25.4%자체 보고

Code

BenchCAD (with Python tool)87.0%자체 보고
BenchCAD67.0%자체 보고
FrontierSWE V243.8%자체 보고

General

GDPval-AA 2.11620.00 / 3000자체 보고
Global-MMLU87.8%자체 보고

Healthcare

HealthBench Professional (raw)71.0%자체 보고
HealthBench Professional64.8%자체 보고
HealthBench61.6%자체 보고
PhysicianBench43.0%자체 보고

Language

MILU87.6%자체 보고

Multimodal

OSWorld 2.1 (partial)72.4%자체 보고
OSWorld 2.1 (strict)37.1%자체 보고

Reasoning

SWE-bench Multilingual83.7%자체 보고
SWE-Bench ProPrinceton NLP (2024)64.8%자체 보고
FrontierCode 1.1 (Extended)58.4%자체 보고
Humanity's Last Exam (with tools)57.4%자체 보고
FrontierCode 1.146.4%자체 보고
Humanity's Last Exam (no tools, text-only)45.9%자체 보고
Terminal-Bench 4.039.2%자체 보고
SWE-Bench Multimodal30.7%자체 보고
Terminal-Bench-Science 0.120.6%자체 보고

Science

LatchBio SingleCellBench56.2%자체 보고

Vision

Chartography86.2%자체 보고
Chartography (no tools)46.4%자체 보고

AA 평가 지수

(Artificial Analysis)
Lcr(Artificial Analysis)
82.7
Scicode(UIUC + Argonne National Lab (2024))
55.0
Hle(Center for AI Safety + Scale AI (2025))
44.4
Intelligence Index(Artificial Analysis)
43.4
Terminalbench V4 0
32.8

LLM Stats 카테고리 점수

(LLM Stats (zeroeval))
Productivity
100
Agents
100
Reasoning
100
General
100
Language
90
Multimodal
60
Science
60
Healthcare
60
Biology
60
Code
60
Vision
60
Math
50
Tool Calling
30

가격

입력 가격$0.1 / 1M 토큰
출력 가격$0.5 / 1M 토큰
혼합 가격 (3:1)$0.2 / 1M 토큰
캐시 읽기 가격$0.01 / 1M 토큰
캐시 쓰기 가격$0.125 / 1M 토큰

속도

토큰/초227.9
첫 토큰 지연292.43s
첫 응답 지연292.43s

공급자 가격 순위

공급자 가격 순위

1개 공급자

공급자입력출력
1Anthropic
$0
$0

이 모델의 다양한 API 공급자 간 가격 비교.

외부 링크