Skip to content
OfferSign up and get $10 of credit for any model in Chat and the API. Top up from $1, with no processing fees. See free models →·Some free models are at the daily cap — refills 05:30 IST; requests auto-route to a live backup. Status →
One AI API Key
Get API key

Anthropic · Claude · Fast

Claude Haiku 5.5

Discount −30%New

Anthropic's fast small model for summaries, classification, coding help and support chat.

Model id

claude-haiku-5-5
Get an API keyTry in Chat

Price per 1M tokens

Input
Regular rate $0.10now $0.07
Output
Regular rate $0.50now $0.35
Cached input
$0.0186% below input

30% below the market rate.

Added here 18 Oct 2026 — read the changelog for what changed and why.

Output is 5× the input rate. $1 covers about 4,065 prompts of 1,000 input + 500 output tokens. 1 credit = $1 of usage. All rates

Start using Claude Haiku 5.5

Set ONEAIAPIKEY_API_KEY to a key from your console, then run any example. Streaming is on so long answers are not cut off by idle-connection timeouts.

curl https://api.oneaiapikey.com/v1/chat/completions \
  -H "Authorization: Bearer $ONEAIAPIKEY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-haiku-5-5",
    "stream": true,
    "messages": [{"role": "user", "content": "Label this support email as billing, bug, feature request or other. Reply with the label only."}]
  }'
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.oneaiapikey.com/v1",
    api_key=os.environ["ONEAIAPIKEY_API_KEY"],
)

stream = client.chat.completions.create(
    model="claude-haiku-5-5",
    messages=[{"role": "user", "content": "Label this support email as billing, bug, feature request or other. Reply with the label only."}],
    stream=True,  # recommended for long outputs
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.oneaiapikey.com/v1",
  apiKey: process.env.ONEAIAPIKEY_API_KEY,
});

const stream = await client.chat.completions.create({
  model: "claude-haiku-5-5",
  messages: [{ role: "user", content: "Label this support email as billing, bug, feature request or other. Reply with the label only." }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}

More: quickstart · API reference · try it in Chat first

Benchmarks reported by the vendor

Scores for Claude Haiku 5.5 as its maker published them, with the setting they state. We have not re-run them. Versions and settings differ between vendors, so compare only identical benchmark names; compare models side by side.

Tokens per day, by model

Last 7 days, UTC days

Hover, tap or drag across any day, or focus the chart and use the arrow keys, to see how every model did that day.

Real tokens served through this gateway, updated every 5 minutes. Ranked by usage, not quality.

Full ranking insights

Claude line-up

Claude Haiku 5.5 or Claude Haiku 4.5 or GLM 5.3 Flash?

Of the 9 Claude models on this site, Claude Haiku 4.5 is the step up from Claude Haiku 5.5, and nothing in the family sits below it. GLM 5.3 Flash from Z.ai is the closest-priced fast-tier model from another vendor. Each comparison lists the published rates and limits, then what favours each side.

Claude Haiku 5.5vsClaude Haiku 4.5

Per 1M tokensClaude Haiku 5.5Claude Haiku 4.5
Input$0.0793% less$1
Output$0.3593% less$5
Cached input$0.0190% less$0.10
Context1M5× larger200k
Max output128k2× more64k

Pick Claude Haiku 5.5 when

  • cost per request decides it: Claude Haiku 5.5 charges input 93% less and output 93% less than Claude Haiku 4.5.
  • prompts run past 200k tokens: Claude Haiku 5.5 reads up to 1M.
  • one answer must be longer than 64k tokens (its limit is 128k).
  • the work is classification and support chat, which this catalogue lists for Claude Haiku 5.5 and not for Claude Haiku 4.5.

Pick Claude Haiku 4.5 instead when

  • the client speaks the Anthropic Messages format, as Claude Code does.
  • your workload looks like extraction and chat, labels the catalogue gives Claude Haiku 4.5 only.

Claude Haiku 5.5vsGLM 5.3 Flash

Per 1M tokensClaude Haiku 5.5GLM 5.3 Flash
Input$0.07the same$0.07
Output$0.3567% more$0.21
Cached input$0.0167% less$0.03

Pick Claude Haiku 5.5 when

  • you rely on Vision, Reasoning and Structured output, which GLM 5.3 Flash does not list.
  • requests include screenshots or other images.
  • the work is classification and support chat, which this catalogue lists for Claude Haiku 5.5 and not for GLM 5.3 Flash.

Pick GLM 5.3 Flash instead when

  • the job runs well enough on the lower rates: output at 40% less.
  • Claude Haiku 5.5 does not list JSON mode and your code depends on it.
  • your workload looks like quick answers and drafting, labels the catalogue gives GLM 5.3 Flash only.

Use cases

What Claude Haiku 5.5 is listed for, and what it costs

The catalogue lists Claude Haiku 5.5 for sub-agents, classification and support chat. Each row prices one typical request at the published rates, with the arithmetic shown. The market-rate figure shows the same request at the lowest public price elsewhere. Token counts are examples; your prompts will differ.

Model your own traffic

  1. Sub-agents

    A helper step for a larger model: 6,000 tokens of file or search results in, a 500-token summary out.

    Route search, summarising and file triage here while a larger model leads.

    “List every function in this file that writes to the database, one per line.”

    Same request on Claude Haiku 4.5 $0.0085 and GLM 5.3 Flash $0.000526. No other fast-tier Claude model on this site does this job for less.

    $0.000596

    per helper call, about 16,778 per $10

    $0.00085 at the market rate

    0.006 × $0.07 + 0.0005 × $0.35

  2. Classification

    One message sorted into a fixed set of labels: 800 tokens in, a 20-token label out.

    Give a fixed label set and ask for one label per item to keep answers short and cheap.

    “Label this support email as billing, bug, feature request or other. Reply with the label only.”

    Same request on Claude Haiku 4.5 $0.0009 and GLM 5.3 Flash $0.00006. No other fast-tier Claude model on this site does this job for less.

    $0.000064

    per message, about 156,250 per $10

    $0.00009 at the market rate

    0.0008 × $0.07 + 0.00002 × $0.35

  3. Support chat

    One chat reply: 1,500 tokens of conversation history in, 400 tokens out.

    “Explain the difference between a mutex and a semaphore in plain words.”

    Same request on Claude Haiku 4.5 $0.0035 and GLM 5.3 Flash $0.00019. No other fast-tier Claude model on this site does this job for less.

    $0.000246

    per reply, about 40,650 per $10

    $0.00035 at the market rate

    0.0015 × $0.07 + 0.0004 × $0.35

Overview

When Claude Haiku 5.5 fits

Claude Haiku 5.5 is a fast, low-cost model from Anthropic, available here under the exact model id “claude-haiku-5-5”. It is the entry point of the Claude line-up here, below Claude Fable 5.1 (free lane), Claude Fable 5.1, Claude Opus 5.5 (free lane), Claude Opus 5.5, Claude Sonnet 5, Claude Sonnet 5.5 (free lane), Claude Sonnet 5.5 and Claude Haiku 4.5. Anthropic has not published a release date we can cite; it has been available here since 18 Oct 2026.

Anthropic documents tools, vision, reasoning, structured output and prompt caching for it, and it reads images as well as text.

It costs $0.07 per million input tokens and $0.35 per million output tokens, so output is 5× the input rate. That is 30% below the market rate of $0.10 input and $0.50 output per million tokens, the lowest public price we found at other API gateways. On the use cases below, output makes up 29% of the bill for sub-agents, 11% for classification and 57% for support chat, so trimming the context you send saves more than capping the answer. With a 3:1 input-to-output mix it ranks 5th lowest-cost of the 27 models on this site.

Cached input bills at $0.01 per million tokens, 86% below the standard input rate, which pays off when the same system prompt or documents go out on every request.

It is served through the OpenAI-compatible format at https://api.oneaiapikey.com/v1, so any SDK or tool that lets you set a custom base URL, for example Codex CLI, Cursor and Cline, can use it. It is not offered on the Anthropic Messages endpoint; pick an Anthropic-format model for Claude Code.

Similar models

Same tier, other vendors

Same key, same balance. Switching is a change to the model field.

Compare side by sideAll models

Use Claude Haiku 5.5 with any balance.

Claude Haiku 5.5 is a paid model: any balance above $0 pays for it, the $10.00 sign-up credit included. When it runs out, add $1 or more to your wallet. Every request is billed at the rates above, with a receipt in your usage log.

Create an accountSee pricing

Claude Haiku 5.5 is developed by Anthropic. Model names and trademarks belong to their owners. One AI API Key is an independent API gateway and is not affiliated with, endorsed by or sponsored by Anthropic.