bedrockavailablesupports_adaptive_thinkingsupports_computer_usesupports_function_callingsupports_max_reasoning_effortsupports_mid_conversation_systemsupports_native_structured_outputsupports_output_configsupports_parallel_tool_use_configsupports_pdf_inputsupports_prompt_cachingsupports_reasoningsupports_response_schemasupports_tool_choicesupports_visionsupports_xhigh_reasoning_effort

Back to Model Explorer

BedrockAvailable on AvalAI

Claude Opus 4.8 API on AvalAI

anthropic.claude-opus-4-8

Claude Opus 4.8 is an AI model from Anthropic that is available through AvalAI's OpenAI-compatible API.

Open dashboardCompare

Price display scale

Input price$5.00 / 1M tokens

Output price$25.00 / 1M tokens

Context window1,000,000 tokens

Maximum output128,000 tokens

Release date2026-05-27

Use Claude Opus 4.8 with the AvalAI API

Keep the API key in an environment variable and send requests from server-side code.

Bash

curl https://api.avalai.ir/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{"model": "anthropic.claude-opus-4-8", "messages": [{"role": "user", "content": "Give a concise, practical solution."}]}'

Python

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["AVALAI_API_KEY"],
    base_url="https://api.avalai.ir/v1",
)

response = client.chat.completions.create(
    model="anthropic.claude-opus-4-8",
    messages=[{"role": "user", "content": "Give a concise, practical solution."}],
)

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

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.AVALAI_API_KEY,
  baseURL: "https://api.avalai.ir/v1",
});

const response = await client.chat.completions.create({
  model: "anthropic.claude-opus-4-8",
  messages: [{ role: "user", content: "Give a concise, practical solution." }],
});

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

Create an API keyRead the quickstart

Capabilities and endpoints

• Adaptive thinking
• Computer use
• Function calling
• Max reasoning effort
• Mid-conversation system messages
• Structured output
• Output config
• Parallel tool use
• PDF input
• Prompt caching
• Reasoning
• Structured output
• Tool choice
• Vision
• Xhigh reasoning effort

Third-party model parameters

Third-party reference data does not change AvalAI endpoint support; use the capabilities and endpoints above.

• include_reasoning
• max_completion_tokens
• max_tokens
• reasoning
• reasoning_effort
• response_format
• stop
• structured_outputs
• temperature
• tool_choice
• tools
• verbosity

Supplemental reasoning effort values

high, low, max, medium, xhigh

Claude Opus 4.8 rate limits

Tier	RPM	TPM
Tier 1	2	30,000
Tier 2	25	450,000
Tier 3	50	800,000
Tier 4	100	1,000,000
Tier 5	150	4,000,000

AvalAI and reference marketplace pricing

Token role	AvalAI final price0% platform fee	OpenRouter cost breakdown
		Model rate	Credit purchase fee	Effective OpenRouter cost after fee
Input price	$5.00 / 1M tokens	$5.00 / 1M tokens	5.5%	$5.28 / 1M tokens
Output price	$25.00 / 1M tokens	$25.00 / 1M tokens	5.5%	$26.38 / 1M tokens

AvalAI pricing is final with a 0% platform fee. The effective OpenRouter prices add its 5.5% credit purchase fee to the highest reported model rate, including long-context tiers. OpenRouter charges a 5.5% fee when credits are purchased ($0.80 minimum), so small credit purchases can have a higher effective cost. OpenRouter FAQ

Peer benchmark performance

Independent, sourced scores with no composite rating.

Artificial Analysis Agentic Index

49.4index

Score	Peer rank	Direction	Source and date
49.4 index	No comparable data	Higher is better	OpenRouter · 2026-08-17

Artificial Analysis Coding Index

74.3index

Score	Peer rank	Direction	Source and date
74.3 index	No comparable data	Higher is better	OpenRouter · 2026-08-17

Artificial Analysis Intelligence Index

57.3index

Score	Peer rank	Direction	Source and date
57.3 index	No comparable data	Higher is better	OpenRouter · 2026-08-17

Third-party model data

Source: OpenRouter

Independent sources supplement this context; AvalAI remains authoritative for pricing, availability, and endpoint support.

Reviewed snapshot 2026-08-17

Provider count

—

Context window range

1,000,000

Maximum output range

128,000

OpenRouter model input rate range

$5 / 1M

OpenRouter model output rate range

$25 / 1M

OpenRouter fee

5.5% credit purchase fee · OpenRouter fee details

Effective input price range after fee

$5.275 / 1M

Effective output price range after fee

$26.375 / 1M

Related models

• Claude Opus 4.7
• anthropic.claude-opus-4-6-v1
• Claude Opus 4.6
• anthropic.claude-opus-4-5-20251101-v1:0

Frequently asked questions

What is the Claude Opus 4.8 API on AvalAI?

Create an AvalAI API key and send the model ID anthropic.claude-opus-4-8 to one of the supported endpoints shown on this page.

How much does the Claude Opus 4.8 API cost on AvalAI?

The Claude Opus 4.8 API on AvalAI costs $5.00 per 1M input tokens and $25.00 per 1M output tokens.

What is the token cost for Claude Opus 4.8?

Current AvalAI pricing for Claude Opus 4.8 is $5.00 / 1M tokens for input and $25.00 / 1M tokens for output.

What is the Claude Opus 4.8 context window?

The recorded maximum input context for Claude Opus 4.8 is 1,000,000 tokens.

What are the rate limits for Claude Opus 4.8 on AvalAI?

On tier 5, Claude Opus 4.8 on AvalAI supports up to 150 requests per minute and 4,000,000 tokens per minute.

What features does Claude Opus 4.8 support on AvalAI?

Claude Opus 4.8 on AvalAI supports Adaptive thinking, Computer use, Function calling, Max reasoning effort, Mid-conversation system messages, Structured output, Output config, Parallel tool use, PDF input, Prompt caching, Reasoning, Structured output, Tool choice, Vision, Xhigh reasoning effort.

What account tier is required to use Claude Opus 4.8 on AvalAI?

Calling Claude Opus 4.8 on AvalAI requires tier 1 or higher.

Third-party data and freshness

• Pricing, availability, endpoints, and rate limits: AvalAI structured data (authoritative)
• OpenRouter marketplace · Last checked: 2026-08-17