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...
Tested on 30 benchmarks · BenchGecko score 69.2. Top scores: Chatbot Arena Elo · Coding (1534.1%), Chatbot Arena Elo · Overall (1474.5%), OTIS Mock AIME 2024-2025 (98.3%).
DeepSeek V4 Flash scores 68.6 (99% as good) at $0.09/1M input · 98% cheaper
Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.
Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.
ARC-AGI 2, harder sequel to ARC. More complex abstract reasoning patterns that test generalization ability beyond training data.
Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.
Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.
- Typemultimodal
- Context1.0M tokens (~500 books)
- ReleasedMay 2026
- LicenseProprietary
- StatusActive
- Cost / Message~$0.035
Frequently Asked Questions
Related Models
phi-3-medium 14B · MicrosoftQwen3.5 397B A17B · Alibaba QwenGPT-5.3-Codex · OpenAIClaude Instant · AnthropicGPT-5.2 Pro · OpenAIKey facts · as of 2026-11-11
- Claude Opus 4.8 by Anthropic. BenchGecko score 69.2, rank 45 of 326 scored models (normalized average of public benchmark scores).
- List price $5.00 input · $25.00 output per 1M tokens (as of 2026-11-11).
- Sold by 11 providers (as of 2026-11-11): Amazon Bedrock $5.00 in / $25.00 out · Anthropic $5.00 in / $25.00 out · Azure $5.00 in / $25.00 out · Claude Platform on AWS $5.00 in / $25.00 out · Google $5.00 in / $25.00 out · and 6 more. Every provider
How to cite · data as of 2026-11-11
Claude Opus 4.8 · benchmarks, pricing and providers. BenchGecko, data as of 2026-11-11. https://benchgecko.ai/model/claude-opus-4-8
Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP