Claude Haiku 5.5
Anthropic's smallest current Claude model, built for high-volume, latency-sensitive work such as classification, routing, extraction and subagent tasks.
- Previous version: Claude Haiku 4.5See what changed
Anthropic's smallest current Claude model, built for high-volume, latency-sensitive work such as classification, routing, extraction and subagent tasks.
Overview
Haiku 5.5 is built for work where speed and cost matter more than depth: sorting and routing requests, pulling fields out of documents and running as a subagent inside larger workflows. It reads text and images and returns text, and connected tools let an application search the web, run code or operate a computer or browser.
Compared with Haiku 4.5 it has a much larger context window, longer outputs, a later knowledge cutoff and lower token prices for shorter prompts; long prompts are billed at a higher rate for the whole request. Anthropic calls it its fastest model at standard speed, which is a provider claim, not a Cody measurement.
Category comparison
These are provider-published specifications, not Cody benchmark scores. Follow the linked sources for current limits and endpoint-specific exceptions.
Pricing & comparisons
Set your usage. Your estimate updates as you type.
Example: 6,000 input tokens for 10 pages, plus a 500-token summary. Page lengths vary; adjust the numbers below.
One run sends your input to the model once and receives an answer. Tokens are pieces of text: input is what you send, output is the answer you receive.
Claude Haiku 5.5
Anthropic
Estimated total (USD)
For the usage above · USD · API pricing, not a subscription
Estimates exclude taxes, tools, cache storage/writes, free allowances and custom discounts. Image estimates cover output only, not prompt or reference-image charges. Quality modes differ by model. Unlisted settings are not treated as free.
API and provider access
Availability
Available through Claude API, Amazon Bedrock, Google Cloud and Microsoft Foundry.
Direct access depends on Anthropic's supported-country list; cloud regions and marketplace eligibility vary.
Check live availabilityData and training
Anthropic does not train on API data without express permission. Retention depends on the API feature, account agreement and cloud provider.
This is a concise reading of the cited provider material, not legal advice. A third-party gateway can have different storage, routing, training, and residency terms from the model maker's direct API.
Read the provider policyFrequently asked questions
Anthropic's smallest current Claude model, built for high-volume, latency-sensitive work such as classification, routing, extraction and subagent tasks. Haiku 5.5 is built for work where speed and cost matter more than depth: sorting and routing requests, pulling fields out of documents and running as a subagent inside larger workflows. It reads text and images and returns text, and connected tools let an application search the web, run code or operate a computer or browser.
Claude Haiku 5.5 was released on October 7, 2026 according to the cited provider materials.
Available through Claude API, Amazon Bedrock, Google Cloud and Microsoft Foundry. The access routes listed in this guide are Anthropic and AWS, Google Cloud, Microsoft.
$0.10 input · $0.50 output / 1M tokens. Rates for prompts up to 100K tokens. Longer prompts are billed at $0.50 input and $2.50 output per 1M tokens for the whole request. Cache reads cost $0.01 per 1M tokens on shorter prompts. Batch costs 50% less; US-only processing adds 10%.
Available through Claude API, Amazon Bedrock, Google Cloud and Microsoft Foundry. Direct access depends on Anthropic's supported-country list; cloud regions and marketplace eligibility vary.
Anthropic does not train on API data without express permission. Retention depends on the API feature, account agreement and cloud provider. The policy belongs to the provider route and account terms, so verify it again before production use.
One price per model, using the same input/output mix. We weight each API rate by its share of one million total tokens and use the lowest matching reviewed route. This is a comparison rate, not a cost per task or a quality score. Tokenizers, caching, reasoning and long context can change your actual bill.
75% input + 25% output · no cache discounts
Anthropic
This model
OpenAI
Mistral AI
OpenAI
OpenAI
Anthropic
OpenAI
Anthropic
OpenAI
Anthropic
OpenAI
Meta
OpenAI
Anthropic
OpenAI
Anthropic
Anthropic
Anthropic
OpenAI
Anthropic
OpenAI
Related comparisons
A Claude model for everyday coding, writing and visual analysis, with room for large reference documents.
Open full comparisonA low-cost OpenAI model for focused tasks repeated at scale, such as extraction, classification and short replies.
Open full comparisonA Claude model for large code changes, detailed research and business tasks that take many steps.
Open full comparisonA compact Claude model for quick replies, classification and narrowly scoped assistants.
Open full comparisonGoogle's stable, high-efficiency multimodal model for agents, software work, and large mixed-media inputs at an introductory Flash-tier price.
Open full comparison