Search operators

brand: benchmark: record: model: metric: date: org:

Separate values with commas, or alternatives with uppercase OR. Example: brand:Meta, Anthropic benchmark:GPQA

Claude Opus 4.5 on ARC-AGI-2

BENCHMARK RECORD: BR-002O45-036

Report record

Canonical URL: https://www.benchmarkregistry.org/records/BR-002O45-036

Benchmark scores may depend on benchmark version, evaluation configuration, model snapshot, and evaluator. Records sharing a benchmark name are not necessarily directly comparable.

Record

Record Identifier
BR-002O45-036
Record Status
ACTIVE
Created At
2026-07-13T13:14:34.963Z
Updated At
2026-07-13T13:14:34.963Z
Superseded By
Not applicable

Model

Official Model Name
Claude Opus 4.5
Model Identifier
ANTHR-O45
Organization
Anthropic
Model Family
Claude
Model Number
45
Tier
O
Model Snapshot
SNP-000004
Snapshot Label
Unknown
Provider Model Identifier
claude-opus-4-5-20251101
Snapshot Date
2025-11-01

Benchmark

Benchmark Family
ARC-AGI
Benchmark Version or Variant
ARC-AGI-2
Benchmark Version Reference
arc-agi/2
Benchmark Organization
ARC Prize Foundation
Benchmark Status
ACTIVE

Result

Metric
Overall
Score
37.6%
Numeric Score
37.64
Evaluation Date
Unknown
Reported Date
2026-07-09

Evaluation Context

Configuration Reference
CFG-000036
Shots
Unknown
Reasoning Mode
reasoning
Reasoning Effort
64k
Pass Count
Unknown
Agent Scaffold
Unknown
Evaluation Harness
Unknown
Temperature
Unknown
Top P
Unknown
Max Output Tokens
Unknown
System Prompt Description
Unknown
Additional Configuration
None reported

Source and Reporting

Primary Source
ARC-AGI Leaderboard
Source Type
LEADERBOARD
Publisher
ARC Prize Foundation
Source Published Date
2026-07-09
Source Accessed At
2026-07-13T12:57:08.196Z
Report Type
BENCHMARK_OWNER
Evaluator
ARC Prize Foundation
Evaluator Type
BENCHMARK_OWNER

Provenance

DateEventSourceDetails
2026-07-13T13:14:34.963ZRecord created manuallyARC-AGI Leaderboard

Record Status

This record is ACTIVE. Registry identifiers remain reserved when record status changes.