Hormouz AIHormouzSign in

Anthropic: Claude Haiku 4.5: AI model

Claude Haiku 4.5 is a fast, cost-efficient model for latency-sensitive and high-throughput applications from Anthropic, with a 200K-token context window. Reachable from inside Iran through Hormouz.

Context window:
200,000 tokens
Input:
$1/1M tokens
Output:
$5/1M tokens
Type:
text, image
  • Vision
  • Tools
  • Streaming
  • Long context
Good for:Popular for Health · #14Ranks are based on global usage data

Thinking levels:

Claude Haiku 4.5 is the fast, compact tier of the Claude family, built for low-latency and high-volume workloads. It accepts text and image input and returns text, supports tool calling for building agents and automations, and works with streaming responses. It also supports extended reasoning, so you can trade extra thinking time for higher accuracy on hard problems. The context window is 200K tokens, which suits long documents, large codebases and multi-turn sessions.

Haiku-class models are commonly used for real-time assistants, high-volume classification and extraction, quick coding help and cost-sensitive pipelines. Hormouz serves it behind an OpenAI-compatible API, so you can point your base URL at Hormouz, drop in a Hormouz key, and run existing OpenAI SDK code without changes.

Pricing is usage-based: $1 per million input tokens and $5 per million output tokens. There is no fixed monthly fee. You pay only for the tokens you use. Hormouz provides stable access to the model from inside Iran, with no foreign credit card or circumvention tooling required.

Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance...

Quickstart

curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-haiku-4.5",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Specs

VendorAnthropic
TypeText, Image
Context window200,000 tokens
Pricing$1.00/M in · $5.00/M out
APIChat Completions
CapabilitiesStreaming · Vision · Tools

Benchmarks

16.9
Intelligence Index
Better than 30% of models we compared
43.9
Coding Index
Better than 46% of models we compared
8
Agentic Index
Better than 35% of models we compared

Reasoning

GPQA Diamond
Graduate-level scientific reasoning
67.2%
HLE
Humanity's Last Exam
10.4%
IFBench
Instruction-following
54.3%
AA-LCR
Long-context reasoning
74.3%
CritPt
Research-level physics reasoning
0%
AA Omniscience
Broad factual knowledge accuracy
18%
Non-Hallucination
Rate of avoiding fabricated answers
72.7%

Coding

SciCode
Python for scientific computing
42.2%
Terminal-Bench Hard
Agentic coding & terminal use
27.3%

Agentic

τ²-Bench
Tool-use agents in dual-control scenarios
54.7%
GDPval
Economically valuable tasks
11.8%

Source: Artificial Analysis

3D
Top 74%
1,083Elo
Win rate: 41.1%Avg time: 29.7s
ASCII Art
Top 61%
1,151Elo
Win rate: 49.3%Avg time: 11.2s
Code
Top 72%
1,121Elo
Win rate: 44.9%Avg time: 48.1s
Data Viz
Top 70%
1,133Elo
Win rate: 45.6%Avg time: 39.2s
Game Dev
Top 69%
1,110Elo
Win rate: 44.7%Avg time: 47.2s
SVG
Top 89%
1,025Elo
Win rate: 39.1%Avg time: 11.6s
UI Component
Top 72%
1,107Elo
Win rate: 42.6%Avg time: 38.0s
Website
Top 70%
1,129Elo
Win rate: 45.1%Avg time: 52.5s

Metrics sourced from Design Arena

Code example

cURL
curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-haiku-4.5",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
Python
from openai import OpenAI
client = OpenAI(api_key="ng-YOUR_API_KEY", base_url="https://api.hormouz.net/v1")
response = client.chat.completions.create(
    model="anthropic/claude-haiku-4.5",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
JavaScript
import OpenAI from "openai";
const client = new OpenAI({ apiKey: "ng-YOUR_API_KEY", baseURL: "https://api.hormouz.net/v1" });
const response = await client.chat.completions.create({
  model: "anthropic/claude-haiku-4.5",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Buy the Anthropic: Claude Haiku 4.5 API, paid in rial

  • Buy credit with an Iranian bank card; no foreign card or account needed.
  • One Hormouz key reaches Anthropic: Claude Haiku 4.5 and 600+ other models, and you pay only for what you use.
  • No VPN and no risk of a banned account; just change the base URL to api.hormouz.net/v1.
Create an accountBuy Claude API access, paid in rial

Use via Hormouz

Connect to this model and 600+ others with one API key, then see usage down to each request.

Request company access

FAQ

How much does Claude Haiku 4.5 cost?
Billing is usage-based: $1 per million input tokens and $5 per million output tokens. There is no fixed monthly fee. You pay only for what you use.
How do I access Claude Haiku 4.5 from inside Iran?
Create an account on Hormouz and generate an API key. Hormouz provides a stable path to the model from inside Iran, with no foreign credit card or circumvention tooling required.
Is Claude Haiku 4.5 compatible with OpenAI code?
Yes. Hormouz exposes an OpenAI-compatible API. Point your base URL at Hormouz and swap in your Hormouz key. Your existing SDKs and code run without a rewrite.
What context window and capabilities does Claude Haiku 4.5 have?
It has a 200K-token context window, accepts text and image input, and supports tool calling and streaming, plus extended reasoning for harder problems. That makes it suitable for long documents, codebases and agentic workflows.
What is Claude Haiku 4.5 good for?
Typical uses are real-time assistants, high-volume classification and extraction, quick coding help and cost-sensitive pipelines.
How does Claude Haiku 4.5 compare to other Claude models?
Claude comes in three tiers: Opus, Sonnet and Haiku. Haiku is the fast, compact tier: the cheapest and lowest-latency option, best for high-volume and real-time work.

Updated 2026-10-13