Claude Opus 4

Tool callingReasoningImage inputFile input

Model card Launch post Website

Flagship Claude model for deep reasoning, coding, and long-horizon agents

Context: 200K
Released: 1y ago

Sources:models.dev

Benchmarks

DeepSWE is the primary coding benchmark where an exact model evaluation exists. Other sourced evaluations appear below; missing scores are not zero.

Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.

Other published benchmarks

Evaluation	Reported score and source
ARC-AGI-2	1.3% Third-party evaluator · claude-opus-4-20250514 External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	8.6% Third-party evaluator · claude-opus-4-20250514_16K Thinking budget: 16K External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	0.0% Third-party evaluator · claude-opus-4-20250514_1K Thinking budget: 1K External evaluation via Epoch AI Original-source terms apply
ARC-AGI-2	4.5% Third-party evaluator · claude-opus-4-20250514_8K Thinking budget: 8K External evaluation via Epoch AI Original-source terms apply
FrontierMath-2025-02-28-Private	4.5% Epoch AI · claude-opus-4-20250514 Epoch AI evaluation
FrontierMath-2025-02-28-Private	4.1% Epoch AI · claude-opus-4-20250514_27K Thinking budget: 27K Epoch AI evaluation
FrontierMath-Tier-4-2025-07-01-Private	0.0% Epoch AI · claude-opus-4-20250514 Epoch AI evaluation
FrontierMath-Tier-4-2025-07-01-Private	4.2% Epoch AI · claude-opus-4-20250514_27K Thinking budget: 27K Epoch AI evaluation
GPQA diamond	69.2% Epoch AI · claude-opus-4-20250514 Epoch AI evaluation
GPQA diamond	76.3% Epoch AI · claude-opus-4-20250514_16K Thinking budget: 16K Epoch AI evaluation
OTIS Mock AIME 2024-2025	42.2% Epoch AI · claude-opus-4-20250514 Epoch AI evaluation
OTIS Mock AIME 2024-2025	60.0% Epoch AI · claude-opus-4-20250514_16K Thinking budget: 16K Epoch AI evaluation
OTIS Mock AIME 2024-2025	64.4% Epoch AI · claude-opus-4-20250514_27K Thinking budget: 27K Epoch AI evaluation
SWE-Bench verified	70.7% Epoch AI · claude-opus-4-20250514 Epoch AI evaluation

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Specifications

Developer: Anthropic
Model family: Claude
Type: Language model
Predecessor: —
Successor: —
Released: May 22, 2025
Model ID: anthropic/claude-opus-4-20250514
Openness: Closed
License: —
Input modalities: text, image, file
Output modalities: text
Modality: text+image+file->text
Max output tokens: 32K
Knowledge cutoff: 2025-03-31
Hugging Face: —
Source freshness: Metadata synced Oct 8, 2026, 10:17 PM UTC

Additional source evidence

Source model: anthropic/claude-opus-4-20250514
Source family: claude-opus
Last updated: 2025-05-22
Knowledge cutoff: 2025-03-31
Model context: 200K
Model output limit: 32K
Open weights: No
File attachments: Yes
Reasoning: Yes
Tool calling: Yes
Temperature control: Yes

Published benchmarks

Benchmark	Score	Evaluation details	Evidence
Aider Polyglot	72 · percent correct	2025-05-25	Source

Complete source record (JSON)

{
  "id": "anthropic/claude-opus-4-20250514",
  "name": "Claude Opus 4",
  "description": "Flagship Claude model for deep reasoning, coding, and long-horizon agents",
  "family": "claude-opus",
  "last_updated": "2025-05-22",
  "knowledge": "2025-03-31",
  "attachment": true,
  "reasoning": true,
  "tool_call": true,
  "temperature": true,
  "modalities": {
    "input": [
      "text",
      "image",
      "pdf"
    ],
    "output": [
      "text"
    ]
  },
  "limit": {
    "context": 200000,
    "output": 32000
  },
  "benchmarks": [
    {
      "name": "Aider Polyglot",
      "score": 72,
      "source": "https://aider.chat/docs/leaderboards/",
      "metric": "percent correct",
      "date": "2025-05-25"
    }
  ],
  "release_date": "2025-05-22",
  "open_weights": false
}

Infrastructure

Provider availability

All providers

Provider endpoint data will appear after this model's first OpenRouter endpoint sync.

Text provenance

Transitioning

Claude text watermarking

Anthropic says Claude models launched before August 2, 2026 have a transition period and are being moved to the same watermarking system over time.

Official explanation