⌕Is ___ SOTA?
isitsota / models / Anthropic / claude-opus-4-8
VERIFIEDOct 132026
Is Claude Opus 4.8 state of the art?
No.Best recorded rank is #1 in General LLM.

As of Oct 13, 2026, Claude Opus 4.8 from Anthropic. Best recorded rank is #1 in General LLM. Last listed at #1 in General LLM on 2026-06-25.

Computed automaticallyas of Oct 13, 2026methodology v1.0confidence: low
General LLMNot SOTA · #1#1 · 0dReasoningNot SOTA · #3#3 · 0dMathNot SOTA · #1#1 · 0dScienceNot SOTA · #29#29 · 0dCodingNot SOTA · #50#50 · 0dCoding · paretoNot SOTA · #50#50 · 0dCoding › AgenticNot SOTA · #14#14 · 0dCoding › Agentic · paretoNot SOTA · #14#14 · 0dCoding › Web devNot SOTA · #1#1 · 0dLong contextNot SOTA · #1#1 · 0d
Alert me if this changesGet badgeCompare with….mdJSON
View as of point-in-time crowns
Why

The ruling, in plain English

Last listed at #1 in General LLM on 2026-06-25.

Receipts

Evidence

Scores come from the latest ingested snapshot
SourceScoreDateTypeReceipt
LMArena Text (style-controlled, overall)
claude-opus-4-8
1479.0558489187351 1473.2–1484.92026-06-25INDEP.snap_51142aff7bb5
LMArena Text (style-controlled, hard prompts)
claude-opus-4-8-high · high
1513.0657764930927 1508.2–1517.92026-09-13INDEP.snap_51142aff7bb5
LMArena Text (style-controlled, math)
claude-opus-4-8
1495.3145155711768 1472.8–1517.92026-06-16INDEP.snap_51142aff7bb5
GPQA Diamond
claude-opus-4-8_max · max
0.9103535353535354 0.9–0.92026-05-28INDEP.snap_f9d8103e52c2
Terminal-Bench 4.0
Opus 4.8 max · max
23.64 20.1–27.22026-10-06INDEP.snap_13678220dcf7
LMArena WebDev
claude-opus-4-8-high · high
1564.2645566436836 1557.0–1571.62026-08-14INDEP.snap_6c327280fd30
LMArena Text (style-controlled, longer query)
claude-opus-4-8-high · high
1502.725890319493 1497.6–1507.92026-10-02INDEP.snap_51142aff7bb5
GDP.pdf
claude-opus-4-8_max · max
0.242026-05-28INDEP.snap_f9d8103e52c2
SimpleQA Verified
claude-opus-4-8_max · max
0.53 0.5–0.62026-05-28INDEP.snap_f9d8103e52c2
ARC-AGI
claude-opus-4-8_max · max
0.9252026-05-28INDEP.snap_f9d8103e52c2
Reign history

Where it sits in the succession

Full timeline →

No closed or open crown interval includes this model.

Honest about the gaps

Where it is not the crown

CategoryThis modelWhy
General LLMNot SOTA · #1Last listed at #1 in General LLM on 2026-06-25.
ReasoningNot SOTA · #3Last listed at #3 in Reasoning on 2026-09-13.
MathNot SOTA · #1Last listed at #1 in Math on 2026-06-16.
ScienceNot SOTA · #29Last listed at #29 in Science on 2026-05-28. That board does not award a crown.
CodingNot SOTA · #50Ranked #50 in Coding.
Coding › AgenticNot SOTA · #14Ranked #14 in Coding › Agentic.
Coding › Web devNot SOTA · #1Last listed at #1 in Coding › Web dev on 2026-08-14.
Long contextNot SOTA · #1Last listed at #1 in Long context on 2026-10-02.