Перейти к основному содержанию

Claude Haiku 5.5 (Max)

AnthropicClaudeProprietary

Описание

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`.

Дата выхода
2026-10-07
Параметры
—
Длина контекста
1.0M
Модальности
image, pdf, text

Радар способностей

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

Рейтинги

Оценки бенчмарков (LLM Stats)

(LLM Stats (zeroeval))

Agents

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

Biology

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

Code

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

General

GDPval-AA 2.11620.00 / 3000Сам.
Global-MMLU87.8%Сам.

Healthcare

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

Language

MILU87.6%Сам.

Multimodal

OSWorld 2.1 (partial)72.4%Сам.
OSWorld 2.1 (strict)37.1%Сам.

Reasoning

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

Science

LatchBio SingleCellBench56.2%Сам.

Vision

Chartography86.2%Сам.
Chartography (no tools)46.4%Сам.

Индексы оценки AA

(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

Оценки категорий LLM Stats

(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

Цены

Цена ввода$0.1 / 1M токенов
Цена вывода$0.5 / 1M токенов
Смешанная цена (3:1)$0.2 / 1M токенов
Цена чтения кэша$0.01 / 1M токенов
Цена записи кэша$0.125 / 1M токенов

Скорость

Токенов/сек227.9
Задержка первого токена292.43s
Время до первого ответа292.43s

Рейтинг цен провайдеров

Рейтинг цен провайдеров

1 провайдеров

ПровайдерВводВывод
1Anthropic
$0
$0

Сравнение цен разных API-провайдеров для этой модели.

Внешние ссылки