Saltar al contenido principal

Claude Haiku 5.5 (Max)

AnthropicClaudeProprietary

Descripción

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`.

Fecha de lanzamiento
2026-10-07
Parámetros
—
Longitud del contexto
1.0M
Modalidades
image, pdf, text

Radar de capacidades

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

Rankings

Dominio#PosiciónPuntuaciónFuente
Capacidad agéntica42
50.0
LS
Ranking de codificación15
93.0
AA
Ranking general23
74.0
AA
Razonamiento matemático39
86.0
LB
Ranking multimodal83
50.0
LS
Razonamiento29
85.0
LB
Ciencia49
76.0
AA

Puntuaciones de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Agents

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

Biology

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

Code

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

General

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

Healthcare

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

Language

MILU87.6%Aut.

Multimodal

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

Reasoning

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

Science

LatchBio SingleCellBench56.2%Aut.

Vision

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

Índices de evaluación 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

Puntuaciones por categoría 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

Precios

Precio de entrada$0.1 / 1M tokens
Precio de salida$0.5 / 1M tokens
Precio mixto (3:1)$0.2 / 1M tokens
Precio de lectura caché$0.01 / 1M tokens
Precio de escritura caché$0.125 / 1M tokens

Velocidad

Tokens/seg227.9
Retraso del primer token292.43s
Tiempo hasta la respuesta292.43s

Ranking de Precios por Proveedor

Ranking de Precios por Proveedor

1 proveedores

ProveedorEntradaSalida
1Anthropic
$0
$0

Comparar precios entre diferentes proveedores de API para este modelo.

Fuentes externas