GPQA Diamond
Graduate-level scientific reasoning
89.8%
ARC-AGI-2
Abstract reasoning on novel puzzles
61.3%
FrontierMath
Expert-level mathematics
79.0%
OTIS Mock AIME
Competition mathematics
98.9%
Google Agent Platform
Claude Haiku 5.5
claude-haiku-5-5 Fast Claude model for responsive assistance, classification, and lightweight agents
VisionTool CallingCachingReasoningWeb
Quick facts
- Released
- October 2026
- Model type
- Proprietary
- Model family
- Claude Haiku
- Input modalities
- Supports: Text, Image, PDF
- Output modalities
- Supports: Text
- Available regions
- USEUGlobal
- Zero data retention
- Yes
- Available on nexos.ai
- Yes
Pricing and specifications of Claude Haiku 5.5
Explore Claude Haiku 5.5 pricing per million input, output, and cached tokens, along with its context window and maximum output limits.
Available providers
Specifications
Context window
1M tokens
Max output
128K tokens
Input
Output
Benchmarks of Claude Haiku 5.5
How Claude Haiku 5.5 scores on independent benchmarks tracked by Epoch AI.
Quality
Epoch Capabilities Index
160.4
ECI rank
#18 of 256
Benchmark data: Epoch AI, ‘Capabilities & benchmarking’, licensed under CC BY 4.0.
Quickstart
Start using Claude Haiku 5.5
Model ID
claude-haiku-5-5 Base URL
https://api.nexos.ai/v1 Code
from openai import OpenAI
# Get your API key from the nexos.ai dashboard
client = OpenAI(
base_url="https://api.nexos.ai/v1",
api_key="YOUR_NEXOS_API_KEY",
)
response = client.chat.completions.create(
model="claude-haiku-5-5",
messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)import OpenAI from 'openai';
// Get your API key from the nexos.ai dashboard
const client = new OpenAI({
baseURL: 'https://api.nexos.ai/v1',
apiKey: 'YOUR_NEXOS_API_KEY',
});
const response = await client.chat.completions.create({
model: 'claude-haiku-5-5',
messages: [{ role: 'user', content: 'Hello!' }],
});
console.log(response.choices[0].message.content);curl https://api.nexos.ai/v1/chat/completions \
-H "Authorization: Bearer YOUR_NEXOS_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "claude-haiku-5-5",
"messages": [{"role": "user", "content": "Hello!"}]
}'Other Anthropic models
Models from Anthropic with comparable capabilities and pricing.
Start building with Claude Haiku 5.5
Access Claude Haiku 5.5 and 200+ other AI models through nexos.ai.
FAQ
Common questions about Claude Haiku 5.5, its availability and usage.
Claude Haiku 5.5 pricing starts at $0.10 per 1 million input tokens and $0.50 per million output tokens. Prices may vary depending on the provider, region, and caching options. See the pricing table above for a full comparison.
Claude Haiku 5.5 supports a context window of 1,000,000 tokens, with a maximum output of 128,000 tokens. The context window determines how much information the model can process in a single request.
Claude Haiku 5.5 supports vision, tool calling, prompt caching, reasoning, and web search. These features can be used individually or together, depending on the model and your application’s requirements.
Claude Haiku 5.5 is available through Google Agent Platform, with deployment options in the US, EU, and Global regions. Availability may vary by provider. See the provider comparison table above for details.
Zero data retention (ZDR) is available for Claude Haiku 5.5 through Google Agent Platform. With ZDR, supported providers do not retain request and response content after processing, subject to their applicable policies.
To use Claude Haiku 5.5 through nexos.ai, you’ll need an Enterprise plan. Once your account is set up, create an API key, set your base URL to https://api.nexos.ai/v1, and specify claude-haiku-5-5 in your API request. nexos.ai is OpenAI-compatible, so you can use your existing SDK integration by updating the API key, base URL, and model ID.