Anthropic

Claude Opus 4

by Anthropictext+image+file->text1 endpoint
Neura Intelligence Index
74.8/ 100
Rank
#59
Confidence
high
Benchmarks
11(28% coverage)

Overview

Claude Opus 4 is benchmarked as the world’s best coding model, at time of release, bringing sustained performance on complex, long-running tasks and agent workflows. It sets new benchmarks in...

Capabilities

Text generation
Image understanding
Tool use / Function calling
Extended reasoning
Prompt caching
Long context
File input
Top-P sampling
Stop sequences

Modalities

Input
ImageTextFile
Output
Text

Technical Specifications

Context Window
200K tokens
Max Output
32K tokens
Knowledge Cutoff
2025-01-31
Tokenizer
Claude
Uptime (24h)
10000.0%
Blended Cost
$30/1M tokens
Country
🇺🇸US
Model Family
Claude 4
Release Date
May 22, 20251y

Supported Parameters

include_reasoningmax_tokensreasoningstoptemperaturetool_choicetoolstop_p

Benchmark Performance

general

BenchmarkScoreSource
Chatbot Arena Elo
1365
aggregated
IFEval
91.5
aggregated
MMLU-Pro
83.5
aggregated

coding

BenchmarkScoreSource
HumanEval
95.1
aggregated
SWE-bench Verified
55.8
aggregated

math

BenchmarkScoreSource
GSM8K
97.8
aggregated
MATH
85.2
aggregated
AIME 2024
35
aggregated

reasoning

BenchmarkScoreSource
BigBench-Hard
93.2
aggregated
GPQA Diamond
74.8
aggregated

multimodal

BenchmarkScoreSource
MMMU
74.2
aggregated

Performance Over Time

Pricing

Per 1M tokens
Input$15.00
Output$75.00
Cache read$1.50
Cache write$18.75
Added
May 22, 2025
Last synced 9/19/2026