> ## Documentation Index
> Fetch the complete documentation index at: https://docs.bytespike.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# claude-opus-4-8

> Anthropic's flagship Claude — codebase-scale reasoning, multi-document synthesis, and hard agentic work with a 200K context window. Extended thinking, web search, and vision all supported. Released 2026-05-29.

`claude-opus-4-8` is the flagship of the Claude family on ByteSpike — Anthropic's most capable model, released **2026-05-29**. It is built for the hardest work: codebase-scale reasoning, multi-document synthesis, long-document review, and agentic tool use. Context window is **200K tokens**.

**Pricing:** $5.00 / 1M input, $25.00 / 1M output, \$0.50 / 1M cache read — see the [rate card](/pricing).

## Protocols

| Protocol                | Path                                                |
| ----------------------- | --------------------------------------------------- |
| Anthropic Messages      | `POST https://llm.bytespike.ai/v1/messages`         |
| OpenAI Chat Completions | `POST https://llm.bytespike.ai/v1/chat/completions` |

## Quickstart

<CodeGroup>
  ```bash cURL theme={null}
  curl https://llm.bytespike.ai/v1/messages \
    -H "x-api-key: $BYTESPIKE_API_KEY" \
    -H "anthropic-version: 2023-06-01" \
    -H "content-type: application/json" \
    -d '{
      "model": "claude-opus-4-8",
      "max_tokens": 2048,
      "thinking": { "type": "enabled", "budget_tokens": 1024 },
      "messages": [
        { "role": "user", "content": "Untangle this contract." }
      ]
    }'
  ```

  ```python Python (anthropic SDK) theme={null}
  import anthropic

  client = anthropic.Anthropic(
      base_url="https://llm.bytespike.ai",
      api_key=os.environ["BYTESPIKE_API_KEY"],
  )

  resp = client.messages.create(
      model="claude-opus-4-8",
      max_tokens=2048,
      thinking={"type": "enabled", "budget_tokens": 1024},
      messages=[{"role": "user", "content": "Untangle this contract."}],
  )

  for block in resp.content:
      if block.type == "text":
          print(block.text)
  ```
</CodeGroup>

## Capabilities

| Capability                      | Supported       |
| ------------------------------- | --------------- |
| Chat completions                | ✅               |
| Streaming (SSE)                 | ✅               |
| Vision (image input)            | ✅               |
| Tool use (function calling)     | ✅ parallel      |
| Prompt caching (cache\_control) | ✅               |
| Extended thinking               | ✅               |
| Web search (web\_search tool)   | ✅               |
| JSON / structured output        | ✅               |
| Context window                  | **200K tokens** |

## When to use

* **Codebase-scale reasoning** — feed a repo into the 200K window, ask for an audit or a multi-file refactor plan.
* **Hard agentic work** — tool use plus extended thinking for multi-step tasks that need a real plan.
* **Multi-document synthesis** — long-document QA and cross-document reasoning where Sonnet falls short.
* **Web search + reasoning** — research-style queries that need both fresh facts and deep analysis.

When **not** to use:

* Default chat at scale — `claude-sonnet-4-6` is one-fifth the cost at the same context window.
* High-volume classification — `claude-haiku-4-5` is far cheaper for simple, repetitive calls.

## Next

* [claude-opus-4-7](/models/claude-opus-4-7) — prior-generation Opus
* [claude-sonnet-4-6](/models/claude-sonnet-4-6) — production mid-tier
* [claude-haiku-4-5](/models/claude-haiku-4-5) — small + cheap