What Is Claude 3.5 Sonnet?

Anthropic released Claude 3.5 Sonnet as the first version in the Claude 3.5 series, providing access at launch through Claude.ai, the Anthropic API, Amazon Bedrock, and Google Cloud Vertex AI.

Subsequently, on October 22, 2024, Anthropic released an upgraded version of Claude 3.5 Sonnet, emphasizing stronger programming capabilities, tool use, and computer-operating abilities. At present, this model is better used as a historical reference model: Anthropic’s deprecation documentation shows that both claude-3-5-sonnet-20240620 and claude-3-5-sonnet-20241022 were retired on October 28, 2025, and it recommends using Claude Sonnet 4.6 as a replacement.

Even so, many users still look up Claude 3.5 Sonnet because, before retirement, the model was widely used in scenarios such as software engineering, document analysis, ultra-long-context prompts, image understanding, and agent-style workflows.

What Are the Main Parameters and Pricing of Claude 3.5 Sonnet?

| Field | Verified value | |-------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------| | Provider | Anthropic (as of June 2026) | | Model series | Claude 3.5 / Claude Sonnet (as of June 2026) | | Model type | Multimodal large language model (as of June 2026) | | Release date | Official announcement date June 21, 2024; upgraded version released on October 22, 2024 (as of June 2026) | | Context window | 200K tokens, based on Anthropic’s official announcement at the time of release (as of June 2026) | | Input pricing | $3.00 per million input tokens, historical launch/API pricing (as of June 2026) | | Cached input pricing | As of June 2026, no official channel has confirmed dedicated pricing for Claude 3.5 Sonnet | | Output pricing | $15.00 per million output tokens, historical launch/API pricing (as of June 2026) | | Billing unit | Per million tokens / MTok, in USD (as of June 2026) | | Supported modalities | Text input, image input, text output (as of June 2026) | | Supported input types | Text and images; the upgraded version’s Beta support for computer use includes screenshots and tools (as of June 2026) | | Supported output types | Text and code (in text form); native image/audio/video output has not been confirmed | | API access | Retired under the Anthropic API on October 28, 2025; historically supported Anthropic API, Amazon Bedrock, and Vertex AI (as of June 2026) | | Model ID | claude-3-5-sonnet-20240620 and claude-3-5-sonnet-20241022, both retired (as of June 2026) | | Availability | Claude API is retired; Anthropic recommends using claude-sonnet-4-6 instead (as of June 2026) | | Knowledge cutoff | As of June 2026, no official channel has confirmed | | Rate limits | As of June 2026, no official channel has confirmed model-specific limits | | Fine-tuning support | As of June 2026, no official channel has confirmed | | Streaming output support | Claude API generally supports streaming output, but it no longer applies after the model’s retirement (as of June 2026) | | Batch API support | As of June 2026, it has not been confirmed whether Claude 3.5 Sonnet still supports the Batch API | | Tool/function calling | In its active period, the upgraded Claude 3.5 Sonnet has been verified to support tool use and Beta computer use (as of June 2026) | | Structured output/JSON mode | As of June 2026, no official channel has confirmed a model-specific JSON mode | | License/usage limits | Proprietary Anthropic model, subject to Anthropic’s service terms; as of June 2026, the model license has not been specified separately |

It should be noted that the current pricing is only for historical reference: in Anthropic’s launch announcement, Claude 3.5 Sonnet was priced at $3 per million input tokens and $15 per million output tokens, and it supported a 200K-token context window. However, the latest deprecation page has marked its API model IDs as retired.

What Core Capabilities Does Claude 3.5 Sonnet Have in Production Environments?

• Long-context document and code analysis

  A 200K-token context window enables Claude 3.5 Sonnet to analyze long documents, multi-file code snippets, and support historical context and technical specifications. However, it’s important to note that a long context window does not guarantee perfect memorization; key outputs still require manual verification.

• Software engineering assistance

  Anthropic highlights improvements in the model for code generation, code editing, code translation, and multi-step software development workflows. The upgraded version released in October 2024 further optimized programming and tool-use capabilities, but before production deployment, testing, code review, and security checks are still required.

• Visual understanding capabilities

  Claude 3.5 Sonnet supports image input and is suitable for parsing charts, screenshots, and text in non-standard images. This gives it some value in document processing and visual quality inspection scenarios, but it cannot replace professional OCR, compliance, or computer-vision systems.

• Agent-style computer operations

  The computer-use public beta introduced with the upgraded version in October 2024 allows developers to instruct Claude to interact with the interface via screenshots, mouse movement, clicking, and input. Anthropic has explicitly stated that this capability is still in an experimental stage and is prone to errors.

When comparing Claude 3.5 Sonnet with other production models, teams typically also evaluate GPT-4o model parameters and API access, Gemini 2.0 Flash multimodal capabilities, and low-cost alternative options such as GPT-4o mini pricing and application scenarios.

What Modalities Does Claude 3.5 Sonnet Support?

| Modality type | Supported? | Description | Source status | |---------------------|--------------------------------------|-------------------------------------------------------------------|-------------------------------------------| | Text input | Supported | Standard prompt and conversation input during the active API period | Verified from Anthropic announcements and API documentation | | Image input | Supported | For visual reasoning, charts, screenshots, and text parsing in non-standard images | Verified from Anthropic announcements | | Audio input | Not confirmed | As of June 2026, there has been no official confirmation that Claude 3.5 Sonnet supports audio input | Not confirmed as of June 2026 | | Video input | Not confirmed | As of June 2026, there has been no official confirmation that Claude 3.5 Sonnet supports video input | Not confirmed as of June 2026 | | Text output | Supported | Main output format | Verified from model behavior and API documentation | | Native image output | Not confirmed | There is no official statement that Claude 3.5 Sonnet is a native image generation model | Not confirmed as of June 2026 | | Audio output | Not confirmed | There is no official statement about dedicated audio output support by the model | Not confirmed as of June 2026 | | Computer interaction | Upgraded Beta capability | Beta support for computer use via API tools; experimental and prone to errors | Verified from Anthropic’s October 2024 announcement |

What Are the Main Limitations of Claude 3.5 Sonnet?

The most important limitation in 2026 is availability: Anthropic has marked the Claude 3.5 Sonnet API model ID as retired, so new production integrations are not recommended to rely on this model. Official recommendations use Claude Sonnet 4.6 as a replacement in its deprecation list.

Claude 3.5 Sonnet also has general AI limitations, such as hallucinations, omitting key information in long contexts, misunderstanding image content, or generating code that looks reasonable but is wrong. These are general AI limitations; unless the provider explicitly states otherwise, they are not model-specific issues.

In computer-use scenarios, Anthropic describes that the October 2024 public beta version was still not fully mature, and operations such as scrolling, drag-and-drop, and zooming presented challenges. Teams using an agent-style workflow need to configure a sandbox, permission controls, monitoring, and human review—especially for sensitive operations.

According to official materials, Claude 3.5 Sonnet is not a native audio, video, or image generation model. For reasoning or large-scale mathematical computation tasks, teams may consider comparing with dedicated reasoning models such as OpenAI o3 or OpenAI o4-mini, but this should be based on specific tasks rather than overall rankings.

What Application Scenarios Is Claude 3.5 Sonnet Best Suited For?

| Application scenario | Reasons it applies | Key limitations | |----------------------------|--------------------------------------------------------|----------------------------------------| | Legacy system maintenance | Helps understand historical systems that integrated claude-3-5-sonnet-20240620 or claude-3-5-sonnet-20241022 | Retired API IDs should be migrated as soon as possible | | Long document summarization | A 200K context window supports large-scale prompt input | Long-context output still requires manual verification | | Code review and migration | Anthropic emphasizes support for programming, code translation, and multi-step development tasks | Code needs testing and human review | | Visual document analysis | Image input supports charts, screenshots, and parsing of text in non-standard images | Does not replace professional OCR/compliance systems | | Agent workflow research | Computer-use Beta has historical significance in GUI agent experiments | Experimental in the public beta stage; prone to errors |

How Does Claude 3.5 Sonnet Compare with GPT-4o and Gemini 2.0 Flash?

| Comparison dimension | Claude 3.5 Sonnet | GPT-4o | Gemini 2.0 Flash | Suitable scenarios | |-----------------------|---------------------------------------------------|---------------------------------------------------|---------------------------------------------------|-----------------------------------| | Provider | Anthropic | OpenAI | Google / Google DeepMind | Depends on the ecosystem | | 2026 model status | Historical model; Anthropic API IDs are retired | Historical model, but still often referenced in comparisons with GPT-4o | Older Gemini 2.x version; subsequent newer Gemini versions exist | Confirm current availability before integration | | Context window | Historically 200K tokens | In the GPT-4o era, API common is 128K tokens | Google states Gemini 2.0 Flash supports 1M-token context | For long-context needs, consider newer active models | | Supported modalities | Text and image input; text output | At launch supported text and visual APIs; GPT-4o’s roadmap is broader | Early documentation for 2.0 Flash supported multimodal input and text output | Choose based on needed input/output modalities | | Programming and agent workflows | Historically strong performance; upgraded version adds Beta computer use | Broad support for multimodal and tool-use workflow | Low-latency multimodal with a focus on native tool use | Test and evaluate for actual tasks | | API status | Anthropic API IDs are retired | Need to consult OpenAI’s latest model list before use | Need to consult Gemini’s latest model list before use | Avoid hard-coding retired model IDs |

This comparison only covers model characteristics that are verified or widely documented, and does not rank which is better or worse. In real production deployments, teams typically should prioritize active successor models rather than Claude 3.5 Sonnet itself.

How to Access Claude 3.5 Sonnet?

As of June 2026, Claude 3.5 Sonnet should be considered a retired Anthropic API model and is not recommended as a target for new integrations. Anthropic lists claude-3-5-sonnet-20240620 and claude-3-5-sonnet-20241022 as retired on October 28, 2025, and recommends claude-sonnet-4-6 as a replacement.

During its active period, Claude 3.5 Sonnet could be accessed via Claude.ai, the Anthropic API, Amazon Bedrock, and Google Cloud Vertex AI. Anthropic’s current API overview shows that the Claude API entry is the POST /v1/messages messages endpoint, and requires authentication headers such as x-api-key or Authorization, anthropic-version, content-type, and so on.

This page does not provide Python or curl executable examples for Claude 3.5 Sonnet, because the verified model IDs have been retired and the related examples no longer apply to the current production environment.

Frequently Asked Questions

• What is the context window size of Claude 3.5 Sonnet?

  According to Anthropic’s launch announcement, Claude 3.5 Sonnet supports a 200K-token context window. As of June 2026, this parameter is only for historical reference because the official has marked the related API model IDs as retired.

• How much does Claude 3.5 Sonnet cost?

  The pricing in Anthropic’s launch announcement is $3 per million input tokens and $15 per million output tokens. Since the API ID has been retired, current production pricing no longer applies.

• Can developers still access Claude 3.5 Sonnet via the Anthropic API?

  Anthropic’s deprecation documentation shows that claude-3-5-sonnet-20240620 and claude-3-5-sonnet-20241022 were retired on October 28, 2025. New integrations should choose an active replacement model.

• What scenarios is Claude 3.5 Sonnet mainly used for?

  Claude 3.5 Sonnet is mainly used for programming assistance, document analysis, long-context reasoning, visual understanding, and early computer-use agent-style workflows. Currently, these uses are mainly for legacy system migration and historical comparisons.