---
title: "Claude Sonnet 4.5"
route_path: "/model/claude-sonnet-4-5-20250929"
canonical_url: "https://www.pipellm.ai/model/claude-sonnet-4-5-20250929"
markdown_path: "/llms/models/claude-sonnet-4-5-20250929.md"
markdown_url: "https://www.pipellm.ai/llms/models/claude-sonnet-4-5-20250929.md"
content_type: "model-detail-page"
description: "Machine-readable detail page for Claude Sonnet 4.5."
generated_at: "2026-10-08T02:59:36.991Z"
---
Canonical page: https://www.pipellm.ai/model/claude-sonnet-4-5-20250929
Markdown mirror: https://www.pipellm.ai/llms/models/claude-sonnet-4-5-20250929.md
Content type: model-detail-page
Generated at: 2026-10-08T02:59:36.991Z
# Claude Sonnet 4.5
## Query Intents
- Understand pricing, provider availability, context window, and capabilities for Claude Sonnet 4.5.
- Compare Claude Sonnet 4.5 against other models available through PipeLLM.
- Find the canonical model identifier to use in SDK or API requests.
## Overview
Claude Sonnet 4.5 is an advanced Anthropic Sonnet model tuned for real-world agents, coding workflows, tool use, and multimodal analysis. It balances frontier-level reasoning with practical latency and cost, making it a strong default for software engineering, automation, and complex business workflows.
## Model Metadata
- Display name: Claude Sonnet 4.5
- Model ID: claude-sonnet-4-5-20250929
- Provider family: Anthropic
- Release date: 2025-09-29T16:01:16.000Z
- Context window: 200K
- Max output: 64K
- Input modalities: text, image, file
- Output modalities: text
- Tool use support: Yes
- Computer use support: Yes
- Cache control support: Yes
## API Integration
Supported call formats through PipeLLM converter routes:
- [OpenAI](https://docs.pipellm.ai/converter/openai-format): `/openai/v1/chat/completions`
- [Responses](https://docs.pipellm.ai/converter/responses-format): `/responses/v1/responses`
- [Anthropic](https://docs.pipellm.ai/converter/anthropic-format): `/anthropic/v1/messages`
- [Gemini](https://docs.pipellm.ai/converter/gemini-format): `/gemini/v1beta/models/{model}:generateContent`

Responses conversion is stateless: use store=false and replay complete output items, including reasoning.encrypted_content, for tool continuations. See the Responses documentation for supported fields and Gemini 3 signature limitations.
## Official Pricing (per 1M tokens)
| Metric | <=200K Context | >200K Context |
| --- | --- | --- |
| Input Price | $3 | — |
| Output Price | $15 | — |
| Cache Read | $0.3 | — |
| Cache Write | $3.75 | — |
| Image Input | $0 | — |
| Image Output | $0 | — |

## Provider Availability
| Provider | Region | Context Window | Max Output | Input Price | Output Price | Cache Read | Cache Write |
| --- | --- | --- | --- | --- | --- | --- | --- |
| AWS | — | 200K | 64K | $3 | $15 | $0.3 | $3.75 |
| GCP Vertex | — | 200K | 64K | $3 | $15 | $0.3 | $3.75 |
| Vertex AI | — | 200K | 64K | $3 | $15 | $0.3 | $3.75 |

## Documentation
Send an authenticated POST request to https://api.pipellm.ai/v1/messages with model set to claude-sonnet-4-5-20250929.