Skip to content

Claude Opus 4.8

ModelActive
by Anthropic · family “claude-opus”

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... (description from the OpenRouter listing)

in: textin: imagein: fileout: textReasoningTool useStructured output
Epoch Capabilities Index
158.3 #10 of 268
90% CI 156.3 – 161.0
Listing ↗
Released
28 May 2026
Context window
1M tokens
Max output
128K tokens
Input $ / 1M tokens
$5.00
Output $ / 1M tokens
$25.0
Cached input $ / 1M
$0.50
Each benchmark shown separately with its own source

Benchmark results (27)

BenchmarkDomainScorevs best recordedSettingRunSource
Furniture Assemblymultimodal42.5% ±6.4
51%
max10 Sep 2026Epoch ↗
LMCAagents57.5%
84%
max—External ↗
DTBenchreasoning94.9%
96%
max—External ↗
Mystery Game Puzzlesgames36.0% ±4.8
43%
max25 Jul 2026Epoch ↗
EBR-benchreasoning28.6% ±1.4
38%
max7 Aug 2026Epoch ↗
OSWorld 2.0agents20.6%
66%
max—External ↗
Surface Evolver Benchscience87.5%
92%
high—External ↗
FrontierMath-Tiers-1-3-v2-Privatemath80.0% ±2.4
85%
max10 Jun 2026Eval log ↗
FrontierMath-Tier-4-v2-Privatemath56.1% ±7.8
57%
max10 Jun 2026Eval log ↗
FrontierCodecoding46.5%
87%
unknown—External ↗
DeepSWEcoding59.0%
80%
max—External ↗
PostTrainBenchagents33.8%
81%
high—External ↗
ProofBenchmath69.0%
69%
max—External ↗
APEX-Agentsagents48.9%
65%
max—External ↗
Chess Puzzlesgames34.0% ±4.8
47%
max29 May 2026Epoch ↗
Remote Labor Indexagents8.3%
40%
unknown—External ↗
SimpleQA Verifiedknowledge53.0% ±1.6
70%
max27 Aug 2026Eval log ↗
FrontierMath-Tier-4-2025-07-01-Privatesupersededmath31.3% ±6.8
65%
max8 Jun 2026Epoch ↗
DeepResearch Benchagents50.2%
91%
high—External ↗
GSO-Benchcoding47.1%
100%
unknown—External ↗
ARC-AGI-2reasoning72.1%
76%
high—External ↗
FrontierMath-2025-02-28-Privatesupersededmath47.2% ±2.9
90%
max8 Jun 2026Epoch ↗
WeirdMLcoding82.9% ±0.0
89%
xhigh—External ↗
OTIS Mock AIME 2024-2025math98.3% ±1.4
98%
max7 Jun 2026Epoch ↗
SimpleBenchreasoning64.8%
79%
unknown—External ↗
GPQA diamondscience91.0% ±1.9
95%
max7 Jun 2026Epoch ↗
ARC-AGIreasoning92.5%
94%
max—External ↗

Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.

API list price over time

Price history

$ per 1M tokens

5 recorded prices on 6 Jun 2026 (OpenRouter listing + Internet Archive snapshots); re-read on every data refresh, most recently 7h ago. Steps show when the price changed.

Events