# Anthropic: claude-haiku-4-5

- Model ID: `claude-haiku-4-5`
- Provider: Anthropic
- Web version: https://www.moleapi.com/en/models/anthropic/claude-haiku-4-5
- Content status: verified

## Model introduction

Claude Haiku 4.5 is Anthropic's compact model for low latency and scaled use, suited to real-time interaction, classification, extraction, and routine automation.

## Model capabilities

- OpenAI compatible
- Anthropic compatible
- Prompt cache
- Reasoning

### Verified specifications

- Official positioning: Fast, compact Claude model
- Context window: 200,000 tokens
- Maximum output: 64,000 tokens
- Input / output: Text, image / Text

## Model pricing and access

Pricing is supplied dynamically by the MoleAPI console API.

### Standard (default, x1)

- Input: $1 / 1M tokens
- Output: $5 / 1M tokens
- Cache read: $0.1 / 1M tokens
- Cache write: $1.25 / 1M tokens

### Discount (discount, x0.8)

- Input: $0.8 / 1M tokens
- Output: $4 / 1M tokens
- Cache read: $0.08 / 1M tokens
- Cache write: $1 / 1M tokens

### Access protocols

| Protocol | Method | Endpoint |
| --- | --- | --- |
| anthropic | POST | /v1/messages |
| openai | POST | /v1/chat/completions |

- Live pricing source: https://home.moleapi.com/api/pricing

## About claude-haiku-4-5

# Claude Haiku 4.5 Model Overview

Claude Haiku 4.5 is a lightweight model in the Claude family released by Anthropic on October 15, 2025. It is positioned as its fastest and most cost-efficient model, offering near-frontier intelligence at a lower price. [7]

## Model Positioning and Core Capabilities

Claude Haiku 4.5 is designed for AI applications sensitive to response speed, with performance in coding, computer use, and agentic tasks that can match Claude Sonnet 4. [7]

Anthropic positions Haiku 4.5 as the fastest latency tier in the Claude family. It supports text and image input, text output, multilingual processing, and vision understanding. [2] The model supports extended thinking, enabling configuration between fast responses and additional reasoning. [2]

## Capabilities and Specifications

Claude Haiku 4.5 provides a 200K-token context window, with a maximum output length of 64K tokens for the synchronous Messages API. [2]

| Specification | Details |
| --- | --- |
| Context window | 200K tokens [2] |
| Maximum output | 64K tokens (synchronous Messages API) [2] |
| Input modalities | Text, images [2] |
| Output modalities | Text [2] |
| Reasoning capabilities | Supports extended thinking; does not support adaptive thinking [2] |
| Reliable knowledge cutoff | February 2025 [2] |
| Training data cutoff | July 2025 [2] |

## Use Cases

Claude Haiku 4.5's low-latency positioning is suited to direct user-facing interaction scenarios such as real-time chatbots and content moderation. [1]

In development and automation workflows, the model can handle coding, computer use, and agentic tasks. [7] B.AI lists it as a suitable model for everyday information-processing tasks such as email classification, meeting summaries, and form data extraction. [1]

## Important Limitations

Claude Haiku 4.5 has a 200K-token context window per request and a 64K-token output limit for synchronous calls. [2]

The model supports extended thinking but does not support adaptive thinking. [2] B.AI describes its complex or multi-step reasoning and creative capabilities as moderate, making it better suited to general tasks and concise information expression. [1]

## Sources

[1] [B.AI Claude Haiku 4.5 模型指南](https://docs.b.ai/llmservice/models/claude-haiku-4-5/)
[2] [Anthropic Claude 模型概览](https://platform.claude.com/docs/en/about-claude/models/overview)
[7] [Claude Haiku 4.5](https://www.anthropic.com/claude/haiku)



## Code examples

### cURL

```bash
curl https://api.moleapi.com/v1/chat/completions \
  -H "Authorization: Bearer $MOLEAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"claude-haiku-4-5","messages":[{"role":"user","content":"Hello"}]}'
```

## More related models

- [claude-fable-5](https://www.moleapi.com/en/models/anthropic/claude-fable-5) — Anthropic
- [claude-opus-5](https://www.moleapi.com/en/models/anthropic/claude-opus-5) — Anthropic
- [claude-sonnet-5](https://www.moleapi.com/en/models/anthropic/claude-sonnet-5) — Anthropic
- [gpt-5.6-luna](https://www.moleapi.com/en/models/openai/gpt-5.6-luna) — OpenAI
- [gpt-5.6-sol](https://www.moleapi.com/en/models/openai/gpt-5.6-sol) — OpenAI
- [gpt-5.6-terra](https://www.moleapi.com/en/models/openai/gpt-5.6-terra) — OpenAI

## Frequently asked questions

### How is claude-haiku-4-5 priced?

Prices are read from the MoleAPI console API and update with model prices, context tiers, and account groups.

### How can I access claude-haiku-4-5?

The catalog currently lists anthropic, openai.

### How do I switch an existing project to claude-haiku-4-5?

Keep the MoleAPI API address and key, replace the model parameter, and check protocol-specific parameters.

### Where do the claude-haiku-4-5 details come from?

Capabilities and limitations are checked against B.AI and the other cited pages; pricing and protocols come from the MoleAPI console.