How do you start calling Claude Opus 4.8?

Use the model ID claude-opus-4-8 through the compatible chat completions endpoint. It supports text, image, and file inputs with text output.

Can you call Claude Opus 4.8 with the SDK you already use?

Yes. Opus 4.8 is served through a compatible endpoint. You can continue using your existing SDK by changing the base URL and model ID.

Does claude-opus-4-8 have a free tier?

Claude Opus 4.8 is billed per use and the record publishes no free-tier allowance. The current published rates are input $5.0000 per 1M tokens and output $25.0000 per 1M tokens. Trial credit or promotional quota is tied to your account rather than to the model, so check the console before planning around a free tier.

When was Claude Opus 4.8 released, and is it still maintained?

Claude Opus 4.8 was released on 2026-05-28. Its catalog status is stable. Check the model page for deprecation notices before you build a long-lived integration on it.

Does Claude Opus 4.8 support tool calling and structured output?

Confirm the current capabilities in the model documentation. Opus 4.8 is designed for autonomous agents. Test with your specific tool definitions.

How do you get a cache hit on claude-opus-4-8?

To get a cache hit on claude-opus-4-8, send requests with identical prompt prefixes. The cache is keyed by the beginning of your prompt; as long as the prefix matches a previous request, subsequent calls benefit from the lower cached rate. Keep system messages and initial context consistent across calls for best results.

What is the max output length for claude-opus-4-8?

claude-opus-4-8 supports up to 128000 output tokens per request. Plan your prompts to stay within this limit, and use the max_tokens or max_completion_tokens parameter to control output length.

Model detail

Claude Opus 4.8

Provided by Anthropic

Pay-as-you-go

Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token context window. It is suited for highly autonomous agents, long-horizon agentic work, knowledge work, and memory-driven tasks where coherence over extended sessions matters. It is particularly strong on multi-step reasoning, complex coding, and end-to-end project orchestration - large codebases, multi-stage debugging, and long-running asynchronous agent pipelines. Beyond coding, it handles knowledge work such as drafting documents, building presentations, and analyzing data, maintaining quality across very long outputs.

Model specs

Model ID: claude-opus-4-8

Context length: 1M
Max output: 128K
I/O modalities
Released: 2026-05

API endpoints

One MixRoute gateway, OpenAI-compatible

Anthropic Messages /v1/messages POST
OpenAI-compatible /v1/chat/completions POST
Gemini /v1beta/models/{model}:generateContent POST

Pricing

Flat rate, unit: /1M Tokens

Input

$5.0000 /1M Tokens

Completion

$25.0000 /1M Tokens

Cache read

$0.5000 /1M Tokens

Cache creation

$6.2500 /1M Tokens

Selection summary

Quickly judge whether it fits your workload

What is this model good for?

Highly autonomous agents and long-horizon knowledge work

Use Claude Opus 4.8 for autonomous agents, memory-driven tasks, and end-to-end project orchestration. It supports text/image/file inputs with a 1M-token context window and maintains coherence across extended sessions.

What should you check before using it?

Test long-horizon agent performance

Opus 4.8 excels at multi-step reasoning, complex coding, and long-running pipelines. Test your multi-stage debugging and asynchronous agent workflows to confirm the model meets your needs. Knowledge cutoff: 2026-01.

Why use it through MixRoute?

Claude Opus 4.8 takes text and images on one route

Text and image inputs share the same route and the same key. MixRoute serves Claude Opus 4.8 on the OpenAI-compatible endpoint /v1/chat/completions; point your existing client at https://api.mixroute.ai/v1 with the model ID claude-opus-4-8 and re-run your integration tests.

Capabilities & compatibility

Unconfirmed capabilities stay "Unknown," no guesses

Tool calling

Claude Opus 4.8 supports function and tool calls.

EX Send a tools definition; the response returns tool_calls to execute and feed results back.

Structured output

Claude Opus 4.8 can produce JSON-formatted output on demand.

EX response_format: { type: "json_schema" } forces valid JSON output.

Streaming

Claude Opus 4.8 supports streamed responses.

EX stream: true streams tokens incrementally, so chat UIs render as they arrive.

Prompt caching

Claude Opus 4.8 supports prompt caching under supported routing conditions.

EX Keep the system prompt and tool definitions at the start of the request; repeated prefixes are automatically billed at the cache-read rate.

Vision input

Claude Opus 4.8 can process image inputs alongside text.

EX Pass a base64-encoded image in a image_url content part alongside text messages.

OpenAI SDK

Call Claude Opus 4.8 through MixRoute's OpenAI-compatible endpoint.

EX Change base_url to https://api.mixroute.ai/v1; everything else follows the OpenAI style.

Token cost estimator

Live estimate from this page's pricing, not an actual bill

Input tokens Max per request 1M

Output tokens Max per request 128K

Requests / month

Cache hit rate 0%

Share of the same prefix read repeatedly, up to 100%

FAQ

Compare claude-opus-4-8 against models of similar capability and pricing on your actual workload. Evaluate latency, output quality, instruction following, and cost per task. Run side-by-side tests with your own prompts — the best model depends on your specific use case, not just benchmark scores.

Get started

01

Create an API key

Generate a key in the Console; one key calls every model.
02

Set the base URL

Fully OpenAI SDK-compatible; just swap the URL to connect.

https://api.mixroute.ai/v1
03

Specify the model ID

In the request's model field, enter:

claude-opus-4-8

Create an API key

New users get $5 in free credits

Test this model and alternate routes with the same request format

Start from a real workload, then let quality, total cost, and failure conditions decide whether to send production traffic.

Create an API key View integration docs