vibecodingnav
中/EN
Models/claude-opus-4-8

Claude Opus 4.8

15th on DeepSWE v1.1 Best (max) at 59.0%. An older row than Opus 5 / 5.5.

Anthropic · Proprietary · Fetched 2026-10-10

LMArena WebDev
1556
#44 / 153 · ±6
The Slop Index
41
#55 / 58
EQ-Bench Creative Writing v3
1840.1
#24 / 146

Benchmark scores

Boards older than 90 days are not counted as recommendations
BenchmarkSettingScoreRankCostAs of
LMArena WebDevhigh1556±6#44—2026-10-27
The Slop Index—41#55—2026-10-28
EQ-Bench Creative Writing v3—1840.1#24—2026-10-29
Convex Evals1 run79.46#12$4.472026-10-28
ReactBenchmax32.329.3–35.3#37$6.772026-10-19
FrontierCodemax · claude-code14.7#21$12.132026-10-27
DeepSWE v1.1max · mini-swe-agent59.057.2–60.7#15$13.222026-09-22
SWE-bench Verified—This board has stopped updating——2026-02-26
Aider Polyglot—This board has stopped updating——2025-11-20

Use it or not

Why
  • Mid-board on DeepSWE, still clearing more than half the tasks.
When not
  • Do not merge this with Opus 5 / 5.5.

Appears in these scenarios

Collected from scenario boards
Build a Web frontend#15DeepSWE 59.0%Build an Android app#15DeepSWE 59.0%Write CLI, scripts, and DevOps#15DeepSWE 59.0%Write docs & copy#24Creative Writing 1840.1

Peers