⚠ This model is no longer available

Claude Haiku 4.5 (20251001) was removed from RealRelay on 2026-10-23. This page is kept as a historical record: prices mentioned below are no longer valid, and the API no longer serves this model.

Claude Haiku 4.5 (20251001)

Claude Haiku 4.5 (20251001) is the dated snapshot of Anthropic's Haiku 4.5 model for fast text and image tasks, with a 200K-token context window and a 64K-token output limit.

Similar models still available

Overview

Claude Haiku 4.5 (20251001) is the dated snapshot of the Haiku 4.5 model. Pinning the snapshot keeps behaviour stable across deployments.

The undated model name resolves to this same snapshot, so the two names point to the same model.

Capabilities

The model supports reasoning, tool calling and image understanding. It accepts text and images and returns text.

Extended thinking is enabled manually, and the model has no effort parameter. The context window is 200K tokens and the output limit is 64K tokens.

Pricing

Input is no longer offered per 1M tokens and output is no longer offered per 1M tokens.

Requests are billed through RealRelay's OpenAI-compatible API.

Use cases

Use the snapshot when an integration needs a fixed model revision.

It handles the same fast text and image tasks as the undated name, with the same context and output limits.

What is Claude Haiku 4.5 (20251001)?

It is the dated snapshot of the Claude Haiku 4.5 model.

Why choose the dated snapshot?

A dated snapshot is a pinned model revision, so behaviour stays stable across deployments.

How much does it cost?

Input is no longer offered per 1M tokens and output is no longer offered per 1M tokens.

What is its context window?

The context window is 200K tokens and the output limit is 64K tokens.

Is it different from the undated Haiku 4.5 name?

No. The undated model name resolves to this same snapshot.

Does it accept images?

Yes. It accepts text and images and returns text.

Pricing

no longer offered
Claude Haiku 4.5 (20251001) was removed from RealRelay on 2026-10-23. This page is kept as a historical record: prices mentioned below are no longer valid, and the API no longer serves this model.

Specifications (as of 2026-10-23)

Context window
200K
Max output
64K
Released
2025-10-01
Knowledge cutoff
Feb 2025
Input
Image / Text
Output
Text
Capabilities
Reasoning, Tool use, Vision