Sulat.com
AI models
$10 off the fastest DeepSeek V4.1 Flash, Kimi K3 and GLM 5.3 from Synthetic
OpenRouter logo

Model details

Claude Haiku Latest

Anthropic's Claude Haiku Latest occupies the efficient tier of the Claude family, designed for developers and teams who need reliable, high-quality AI without the compute overhead of larger models. It always redirects to the most current Haiku variant, meaning users automatically benefit from the latest improvements Anthropic rolls out to this model line. This living design keeps the model fresh without requiring migration or re-evaluation of endpoints. Built with vision capabilities alongside text processing, Haiku Latest serves applications ranging from document analysis to conversational assistants where broad knowledge and good reasoning matter more than sheer scale.

The Haiku family has earned a reputation for balanced performance—fast response times with quality that holds up across coding, writing, research, and creative tasks. The model supports tool calling, structured output, and function calling, making it practical for building agents and workflows that require precise, machine-readable responses. Its broad knowledge base and reasoning abilities mean it handles multi-step tasks and nuanced prompts without frequent hand-holding. Anthropic's positioning of Haiku as the accessible entry point in their Claude lineup reflects a deliberate choice to serve high-volume, cost-sensitive use cases where the efficient tier delivers the right trade-off between capability and cost.

OpenRouter~anthropic/claude-haiku-latestclaude-haiku

Quick Info

Powered by
Provider
OpenRouter
Model key
~anthropic/claude-haiku-latest
Release date
Apr 27, 2026
Last updated
Apr 27, 2026
Input modalities
Output modalities
Capabilities

Cost

Input token cost
$1.00
Output token cost
$5.00

Limits

Output tokens
64,000 tokens
Context window
200,000 tokens

Transparent token rates

Compare Claude Haiku Latest pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about Claude Haiku Latest

OpenRouter

CoverageBenchmark

A third-party benchmark page from OpenOCR explicitly evaluates "Anthropic Claude Haiku Latest" using the OpenRouter provider offering openrouter--anthropic-claude-haiku-latest, achieving an average score of 98.2% across 11 documents in the corpus ocr-shared-11-normalized/v2, with all scores reproducible (11/11) and ave Per-document results on the page include strong OCR performance on receipts (grocery receipt 99.6%, café receipt 99.3%), handwriting (handwritten letter 100%, handwritten notes 100%), and other document types, with character LCS similarity used as the scoring metric. The evidence is narrowly scoped to OCR capability on

Videos about Claude Haiku Latest

More models around Claude Haiku Latest