Your first $5 becomes $15Get started
Back to Models
Anthropic

Claude Opus 4

Claude Opus 4 from Anthropic — 200K-token context, multimodal vision, function calling.

Released

Try in Playground
Context Window

200.0K

tokens

How much text the model can consider at once — your prompt plus its reply.

Max Output

8K

tokens

The most tokens the model can generate in a single response.

Input

$15.00

per 1M tokens

Cost per 1 million input tokens you send to the model.

Output

$75.00

per 1M tokens

Cost per 1 million output tokens the model generates.

Provider & Routing

Maker
Anthropic
Served on
Anthropic

NemoRouter is a managed gateway — one API key, no provider accounts to set up. We route your traffic to this model's hosting cloud and fail over automatically if it degrades.

Capabilities

Vision
Function Calling
Streaming
JSON Mode
System Prompt
Tool Use

More from Anthropic

Quick Start

curl
# Cache: enabled (org default). Pass nemo_cache: false to skip.
# Use -i to see cost + guardrails in response headers:
#   x-nemo-request-cost: 0.00347
#   x-nemo-guardrails-applied: pii-detection,prompt-injection
# nemo_cache: pass "nemo_cache": false in body to skip cache
curl https://api.nemorouter.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $NEMOROUTER_API_KEY" \
  -d '{
    "model": "claude-opus-4-20250514",
    "temperature": 1,
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Hello! What models do you support?"}
    ]
  }'

Replace sk-nemo-... with your NemoRouter API key. Use claude-opus-4-20250514 as the model identifier.

Try in Playground

Paste your NemoRouter virtual key and send a real request to Claude Opus 4 right now — no signup required.

Test it

Prompt
cURL
curl https://api.nemorouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $NEMOROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"claude-opus-4-20250514","messages":[{"role":"user","content":"Hello!"}]}'