Claude Opus 4.6

Name: Claude Opus 4.6 API
Brand: Anthropic
SKU: claude-opus-4-6

silvamux/claude-opus-4-6

Claude Opus 4.6 is Anthropic's model for demanding tasks, handling complex codebases, in-depth analysis and multi-step agentic workflows — suited to engineering, research and professional content production.

Supported modalities

text

Input price$5 / 1M tokensper 1M tokens

Output price$25 / 1M tokensper 1M tokens

Pricing

Tune and Pro share the same prices.

Input context	Input	Output	Cache read	Cache write
-	$5 / 1M tokens	$25 / 1M tokens	$0.5 / 1M tokens	$6.25 / 1M tokens

Capabilities

Claude Opus 4.6 scored 78 overall, performing well — a knowledge-oriented generalist. It stands out in general knowledge (94%), hallucination control (91%) and multilingual capability (87%), combining broad accurate knowledge with dependable cross-language output, which serves customer support, localization and content moderation well. Comparatively, code generation (58%) is weak; coding tasks are not its strength. For multilingual, accuracy-sensitive work it remains a solid choice.

Overall

78/ 100良好

HighlightsGeneral Knowledge 94%Hallucination Control 91% · Multilingual Capability 87%

Best forCustomer Support / Translation & Localization / Content Moderation & Risk ControlScenario fit 83%

Relative weaknessCode Generation58%

Advanced Capabilities

8 items

Agent Collaboration74%

18 questions · Mixed

Code Generation58%

12 questions · Objective

Code Comprehension & Debugging79%

6 questions · Mixed

Creative Writing63%

22 questions · Mixed

Information Retrieval / RAG73%

9 questions · Mixed

Long-Context Handling74%

16 questions · Mixed

Math & Science Reasoning83%

14 questions · Mixed

Multilingual Capability87%

11 questions · Subjective

General Foundations

3 items

General Knowledge94%

10 questions · Mixed

Text Comprehension85%

23 questions · Subjective

Logical Reasoning68%

16 questions · Mixed

Safety & Compliance

2 items

Hallucination Control91%

15 questions · Subjective

Instruction Following82%

18 questions · Subjective

Performance

Claude Opus 4.6

claude-opus-4-6@260909

Last 24 hours · 54,363 requests

First token · TTFT · P50: 308ms
Output speed · TPS · P50: 49tok/s
Full response · E2E · P50: 5.38s
Request success rate: 98.03%

24-hour availability

24 hours ago12 hours agoNow

Claude Opus 4.6 example code and API

Proclaude-opus-4-6@260909

/api/v0/chat/completions/api/anthropic/v1/messages

UnifiedAnthropic

curl https://silvamux.io/api/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-opus-4-6@260909",
    "messages": [{"role": "user", "content": "Hello, please introduce yourself"}]
  }'

# pip install openai
from openai import OpenAI

client = OpenAI(
    base_url="https://silvamux.io/api/v1",
    api_key="YOUR_API_KEY",
)

response = client.chat.completions.create(
    model="claude-opus-4-6@260909",
    messages=[
        {"role": "user", "content": "Hello, please introduce yourself"}
    ],
)

print(response.choices[0].message.content)

// Gradle: implementation(__HL_SLOT_0__)
import okhttp3.*;

OkHttpClient client = new OkHttpClient();
MediaType json = MediaType.get("application/json");
String payload = "{\"model\":\"claude-opus-4-6@260909\",\"messages\":[{\"role\":\"user\",\"content\":\"Hello, please introduce yourself\"}]}";

Request request = new Request.Builder()
    .url("https://silvamux.io/api/v1/chat/completions")
    .header("Authorization", "Bearer YOUR_API_KEY")
    .post(RequestBody.create(payload, json))
    .build();

try (Response response = client.newCall(request).execute()) {
    System.out.println(response.body().string());
}

// npm install openai
import OpenAI from 'openai'

const client = new OpenAI({
  baseURL: 'https://silvamux.io/api/v1',
  apiKey: 'YOUR_API_KEY',
})

const response = await client.chat.completions.create({
  model: 'claude-opus-4-6@260909',
  messages: [{ role: 'user', content: 'Hello, please introduce yourself' }],
})

console.log(response.choices[0].message.content)