AnthropicUnited States· released May 2026

Claude Opus 4.8

Name: Claude Opus 4.8
Author: Anthropic

Ranked #9, 20.5% behind the frontier on average.

Anthropic, United States. One of 12 models from this lab on the Index.

Third-party publishedHigh confidence (12+ comparable tests, all 4 Index categories)

Available in Theo as Theo Code

Theo orchestrates it for the steps it does best and always shows which engine answered.

Use it in Theo

Snapshot September 8, 2026

OpenCharts Index

79.5

Ranked #9 (7 to 9) of 36 · 20.5% behind the frontier on average · 21 comparable tests · 4 counting categories

Drop any single test and the Index lands between 77.7 and 81.8, anywhere from #7 to #9. Neighbours inside that range are ties.

Needs more results (3 of 8 so far)

Averaging 72.9 on the 3 it has; the score opens at 4 of 8.

Distance to today's frontier on the hardest tests. Not a measure of intelligence.

What it's known for

Claude Opus 4.8, in 21 results.

Strongest in reasoning (81.4) and agentic (81.2). It trails the frontier most in math, 22.4% behind on average. No comparable result yet in long context. Between 20% and 25% behind the best published results, on average. Drop any single test and it would sit anywhere from #7 to #9.

Strongest category

Reasoning

Hard, novel problems: graduate science, abstract puzzles, expert exams.

81.4

ARC-AGI-192.5% · 94% of best
GPQA Diamond91.0% · 93% of best
SimpleBench64.8% · 79% of best
ARC-AGI-272.1% · 76% of best
CritPt20.9% · 65% of best

Mean of 5 of the category's 6 comparable tests. Counts toward the Index.

Best single result

Mock AIME 2024–2025

Solve competition mathematics problems that have a single exact integer answer.

98.3%

98.3% of the best (100.0%) · #10 of 78.

Olympiad-style problems need long, careful chains of reasoning, and there is no partial credit.

Where it trails

Code Arena (WebDev)

Build a web app from the same prompt as a rival model; real users vote blind on the result.

1,563

58.8 points short of parity with the board leader (1,797) · #19 of 86.

People judging finished apps side by side is the most honest measure of front-end quality.

Not measured yet

No comparable result yet in this category. A blank is a blank, never a zero, and it does not lower the Index.

Long context

Covered, but not averaged into the Index:

Knowledgesingle-test category, shown beside the Index · 70.1
Multimodalsingle-test category, shown beside the Index · 92.0
Human preferencesingle-test category, shown beside the Index · 92.9

Every score, with its source

See it at work

What Claude Opus 4.8 was asked to do, and how it did.

Every test it has taken, by category, with the kind of task it faced, who produced the number and where the result landed against the best published one. The examples are original and illustrative, never items from the datasets themselves.

Human preference · mean 92.9 across 1 comparable test · a single-test category, shown beside the Index and never averaged into it

Text ArenaLeaderboard

Answer the same prompt as a rival model; real users vote blind on which answer they prefer.

1,482

46% expected win rate against the board leader (1,507)
#15 of 92

high · board 2026-09-02

Source: Arena

Every result is read against the best published one on its test: percent scores as a share of the best above chance, Arena ratings as the expected win rate against the board leader, open-ended values on a log scale. Sitting a harder exam never lowers a model. Every number links to the publisher that produced it.

Head to head

Claude Opus 4.8 against whoever you pick.

Choose a rival. Every counted test both have taken appears side by side as a share of the best published result on that test, with the category means above. A win is a higher share; a gap under 2 points is a tie, because one item on a 45-problem exam is 2.2 points and publishers' own standard errors are of that order.

Claude Opus 4.8

#9·Index 79.5B

0 : 13

wins · 1 tie · 14 shared

GPT-6 Astra

#1·Index 99.0A+

GPT-6 Astra comes out ahead on 13 of the 14 tests both have taken.

Category means

Reasoning
81.4
99.6
Knowledge
70.1
100.0
Coding
77.7
97.7
Math
77.6
99.8
Agentic
81.2
98.5
Multimodal
92.0
—
Human preference
92.9
—

Shared tests · biggest gaps first

Code Arena (WebDev)Coding
1,563
1,797
−58.8
FrontierMath Tier 4Math
56.1%
97.6%
−42.5
CritPtReasoning
20.9%
31.7%
−33.6
ProofBenchMath
69.0%
99.0%
−30.0
SimpleQA VerifiedKnowledge
53.0%
75.6%
−29.9
ARC-AGI-2Reasoning
72.1%
95.0%
−24.1
FrontierMath (Tiers 1–3)Math
80.0%
93.7%
−14.6
FrontierCodeCoding
46.5%
53.3%
−12.7
WeirdMLCoding
82.9%
92.9%
−10.8
APEX-AgentsAgentic
42.5%
46.7%
−8.8
GPQA DiamondReasoning
91.0%
95.8%
−6.7
ARC-AGI-1Reasoning
92.5%
98.5%
−6.1
SciCodeCoding
53.5%
56.5%
−4.8
Mock AIME 2024–2025Math
98.3%
100.0%
tie

Bars are distances to the best published result on each test (percent scores above chance, Arena ratings as win rate against the leader, open-ended values on a log scale). Raw values as the publishers report them.

Open GPT-6 Astra

Scores

Every test, every source

The best published run per benchmark, the raw value as the publisher reports it, who produced the number, how close it comes to the best published result on that test, and where the model sits among every model scored on it. Percent scores are read above chance, Arena ratings as a win rate against the board leader, open-ended values on a log scale. A test with too few models to compare against is shown but not counted.

Category means

Reasoning81.4 · 5/6
Knowledgebeside70.1 · 1/1
Coding77.7 · 5/7
Math77.6 · 4/5
Agentic81.2 · 4/8
Multimodalbeside92.0 · 1/1
Human preferencebeside92.9 · 1/1
Long context—

Mean · tests taken of the category's comparable tests. A faded category either rests on fewer than half of them or is a single-test category (marked “beside”), which is shown next to the Index and never averaged into it.

GPQA DiamondEpoch-runReasoning

91.0%

93.3% of the best (95.8%), above chance
74th percentile of 78 · n = 198

max

Claude Opus 4.8

Claude Opus 4.8, in 21 results.

What Claude Opus 4.8 was asked to do, and how it did.

Claude Opus 4.8 against whoever you pick.

Every test, every source

Compare with its neighbours.

The best models, ranked here, working inside Theo.