Claude 4 Models by Anthropic
5 models2025Up to 200k ctxFrom $3/1M input
About
Claude 4 family from Anthropic, introducing improved reasoning and capabilities.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
2 in view3 retired
Claude Opus 4.1Current
Use when the workload needs 200k context and structured outputs.
2025-08200k contextstructured outputs
Claude 4 HaikuCurrent
Use when the workload needs 200k context, tool use, and function calling.
2025-05200k contexttool usefunction calling
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Claude Opus 4.1 | Use when the workload needs 200k context and structured outputs. | 2025-08 | 200k contextstructured outputs | Current |
| Claude 4 Haiku | Use when the workload needs 200k context, tool use, and function calling. | 2025-05 | 200k contexttool usefunction calling | Current |
Release Timeline
2 release groups2025-08
1 current · 1 retired
Claude Opus 4.1
Current200k contextstructured outputs
Claude Sonnet 4
Replaced200k contextstructured outputsmultimodal inputs
2025-05
1 current · 2 retired
Claude 4 Haiku
Current200k contexttool usefunction calling
Claude 4 Opus
Replaced200k contexttool usefunction calling
Claude 4 Sonnet
Replaced200k contexttool usefunction calling
Replaced By
Replaced
Keep for legacy integrations; evaluate Claude Sonnet 4.6 before new work.
Replaced
Keep for legacy integrations; evaluate Claude Sonnet 4.6 before new work.
Replaced
Keep for legacy integrations; evaluate Claude Opus 4.7 before new work.
Specifications(5 models)
| Model | Released | Context | Vision | Multimodal | Fn Calling | Tool Use | Structured Outputs |
|---|---|---|---|---|---|---|---|
| Claude Opus 4.1 | 2025-08 | 200k | No | No | No | No | Yes |
| Claude 4 Haiku | 2025-05 | 200k | Yes | Yes | Yes | Yes | No |
Available From(7 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| Claude Opus 4.1 | Anthropic | $15 | $75 | Serverless |
| Claude Opus 4.1 | AWS Bedrock | $15 | $75 | Serverless |
| Claude Opus 4.1 | OpenRouter | $15 | $75 | Serverless |
| Claude Opus 4.1 | Microsoft Foundry | $15 | $75 | Serverless |
| Claude Opus 4.1 | Vercel AI Gateway | $15 | $75 | Serverless |
Frequently Asked Questions
- What is Claude 4 used for?
- Claude 4 is used for vision and multimodal work, agent workflows and tool use, and structured outputs. The family description and listed model capabilities point to those workloads as the best fit.
- How does Claude 4 compare to Computer Use?
- Claude 4 by Anthropic is strongest where you need vision and multimodal work, while Computer Use by Anthropic is the closest related family to check for agent. Claude 4 has 5 listed variants and reaches up to 200k context, so compare the specs and pricing tables before choosing a production model.
- Which Claude 4 model should I use?
- For the lowest listed input price, start with Claude Sonnet 4 through Anthropic at $3/1M input tokens. For the most capable/latest local choice, evaluate Claude 4 Haiku with 200k context and tool use, function calling, and multimodal inputs.
