Claude Haiku 5.5 Price: API Costs $0.10 Input, $0.50 Output per Million Tokens

Price data as of 2026-10-09. Figures come from the BazaarLink model list. The model page always shows the current value.

Claude Haiku 5.5 costs $0.10 per million input tokens and $0.50 per million output tokens on the BazaarLink model list, with a 1,000,000-token context window. Per Anthropic's announcement, prompts above 100,000 tokens are priced at $0.50 input and $2.50 output. Prices are in US dollars.

Claude Haiku 5.5 price table

The model id is anthropic/claude-haiku-5.5. BazaarLink bills at the list price with no markup. Anthropic's announcement lists the tiered prices above and cache read and write prices. Actual BazaarLink billing follows the model page and your invoice.

Price compared with the previous generation and similar small models

This compares published list prices only and ranks nothing on capability. Base rates, below each long-prompt threshold:

The Haiku 4.5 figure comes from Anthropic's announcement and may not appear in the BazaarLink list. Haiku 5.5 and GPT 6 Luna share the same $0.10 and $0.50 base rate; the difference is long prompts. Haiku 5.5 moves to $0.50 and $2.50 above 100,000 tokens per Anthropic, while Luna moves to $0.20 and $0.75 above 272,000. Anthropic says Haiku 5.5 is best value for prompts up to 100K, which it says made up about 90% of requests to the previous Haiku (vendor figure). For the upgrade trade-offs, see Haiku 4.5 vs 5.5.

Estimating cost

Illustrative arithmetic, not measurements:

• Short prompts, high volume: 1,000 requests a day, 10,000 input and 1,000 output tokens each, 30 days, so 300M input and 30M output tokens. 300 x $0.10 + 30 x $0.50 = $45 a month. The same workload on Sonnet 5.5 is $900.
• One long prompt (estimated with the long-prompt rates in Anthropic's announcement): 150,000 input and 2,000 output tokens, assuming the whole request uses the long-prompt rates: 150,000 x $0.50 per million + 2,000 x $2.50 per million = $0.075 + $0.005 = $0.08. The same request on GPT 6 Luna (below its 272,000 threshold) is about $0.016.

Haiku 5.5 suits high-frequency small tasks under 100K tokens. If your prompts regularly pass 100K, consider chunking or summarizing before you send.

How to call it with the OpenAI SDK

Point base_url at BazaarLink and keep the key in an environment variable named BAZAARLINK_API_KEY, never in source.

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://bazaarlink.ai/api/v1",
    api_key=os.environ["BAZAARLINK_API_KEY"],
)

resp = client.chat.completions.create(
    model="anthropic/claude-haiku-5.5",
    messages=[{"role": "user", "content": "Classify this support chat as refund, shipping, technical or other."}],
)
print(resp.choices[0].message.content)

In Node.js, set baseURL to the same address, read process.env.BAZAARLINK_API_KEY, and pass anthropic/claude-haiku-5.5 as the model.

Before you commit to a budget

A price table only gets you to a rough number. Four checks make the estimate much closer to what you will actually pay:

1. Find your 100K line. Anthropic's announcement lists a higher rate above 100,000 prompt tokens, so check what share of your requests crosses it. Chat history and attached files are the usual cause.
2. Read the usage field. Every response reports input and output token counts. Send a sample of 100 real requests, sum the usage, and multiply by the rates above.
3. Test quality on your own labels. Classification, extraction and summarizing are typical small-model jobs, so run a labeled sample and measure accuracy rather than relying on a benchmark table.
4. Compare with a neighbor. Run the same sample on GPT 6 Luna, which shares the base rate, and on Sonnet 5.5 to see what a higher tier buys you for your specific task.

Moving between these models only means changing the model id in your request.

FAQ

How much does the Claude Haiku 5.5 API cost?

The BazaarLink model list (2026-10-09) shows $0.10 input and $0.50 output per million tokens. Per Anthropic's announcement, prompts above 100,000 tokens are $0.50 input and $2.50 output. Actual BazaarLink billing follows the model page and your invoice.

What is the Haiku 5.5 context window?

1,000,000 tokens. It accepts text, image and file input and returns text.

How much cheaper is Haiku 5.5 than Haiku 4.5?

Per Anthropic's announcement, Haiku 4.5 is $1 input and $5 output per million tokens. Haiku 5.5 is one tenth of that up to 100K prompts; Anthropic says the average real cost is about 75% lower.

Is Haiku 5.5 priced the same as GPT 6 Luna?

The base rate is the same ($0.10 input, $0.50 output), but the long-prompt threshold and rates differ. Per Anthropic, Haiku 5.5 steps up above 100K, and Luna above 272K on the BazaarLink list, so the better deal depends on your prompt length.

Does BazaarLink add a markup on Haiku 5.5?

No. It bills at the model's list price. Prices are in US dollars, and Taiwan-based customers can pay in NTD and receive a Uniform Invoice.

Will this price change?

It can. The data is dated 2026-10-09, so check the model page before you rely on it.

Related

• Haiku 5.5 model page: /en/models/anthropic-claude-haiku-5.5
• All models and live prices: /en/models
• Similar models: Haiku 5.5 vs other small models, GPT 6 Luna API pricing
• Vendor source: Anthropic's Haiku 5.5 announcement

TWD billing · Taiwan invoices · leading AI models · OpenAI-compatible API