Claude Haiku 5.5
Claude Haiku 5.5: the fastest, cheapest fifth-generation Claude for high-volume tasks and sub-agents, 1M context.
First request
from openai import OpenAI
client = OpenAI(
base_url="https://api.altrouter.ai/v1",
api_key="ar-...",
)
resp = client.chat.completions.create(
model="claude-haiku-5.5",
messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.altrouter.ai/v1",
apiKey: process.env.ALTROUTER_API_KEY, // ar-...
});
const resp = await client.chat.completions.create({
model: "claude-haiku-5.5",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(resp.choices[0].message.content);curl https://api.altrouter.ai/v1/chat/completions \
-H "Authorization: Bearer ar-..." \
-H "Content-Type: application/json" \
-d '{
"model": "claude-haiku-5.5",
"messages": [
{
"role": "user",
"content": "Hello!"
}
]
}'The interface is OpenAI-compatible: in existing code only the base_url, the key and the model id change — nothing else.
Claude Haiku 5.5 against its price neighbours
| Model | Input / 1M | Output / 1M | Context | |
|---|---|---|---|---|
| GPT-6 LunaOpenAI | $0.085 | $0.42 | 1.1M | Open |
| Claude Haiku 5.5this pageAnthropic | $0.085 | $0.42 | 1M | |
| GPT-4o MiniOpenAI | $0.13 | $0.51 | 128K | Open |
| GPT-5.6 LunaOpenAI | $0.17 | $1.02 | 400K | Open |
| GPT-5.4 nanoOpenAI | $0.17 | $1.06 | 400K | Open |
Prices and parameters come from the same catalog as the price above; this is not a benchmark, it is comparable billing terms.
Limits
Context is 1M tokens per request, and it holds everything you send: the system prompt, the conversation history and the current question together. Whatever does not fit is not seen by the model — trimming the history is your side of the job, and you pay for exactly what fit.
Response length is set by max_tokens, capped at 8,192 tokens per call. A response that hits the cap comes back with finish_reason: "length" — truncated but still billed, so it is worth setting deliberately on long generations.
Supported: a reasoning mode, tool calling, image input.
There is no subscription: every successful request is billed at the price above, and a failed one is not billed at all. That price is already 15% below the vendor's official one — the official price is shown struck through next to it so it can be checked. No Anthropic account of your own is needed: requests go to api.altrouter.ai and we make the upstream call.
Rate limiting is shared across models and counted per key: 600 requests per minute by default, and a 429 with retry-after: 60 above it. An individual key can get its own limit and a spending cap per day, week, month or lifetime in the dashboard.
When to pick Claude Haiku 5.5, and when a neighbour
The dearer neighbour is GPT-6 Luna: $0.085 per 1M input tokens. What the catalog says you get for it: 1.1M of context instead of 1M. If your task needs none of those, the premium buys nothing.
Guessing is more expensive than checking: the table above is prices and parameters, not a benchmark, and it cannot tell you which model does better on your own prompt. Switching costs one line — the model id in the request changes, the base_url and the key stay — so running both on your own data takes minutes.
OpenAI-compatible chat completions endpoint, with SSE streaming.
/v1/chat/completionsStructured fields (tools, response_format, stop and more) are in the full parameter reference.