AnthropicProvisional

Claude Opus 4.8

Claude 4.8 / Opus · claude-opus-4-8

Reliable complex agentic coding and enterprise engineering model.

Objective score

35.6

Confidence

79%

Price evidence

$5 input / $25 output per 1M tokens

Support this model Compare trend Official source

Scenario evidence

Six task-specific rankings

Feature development#33Obj. 35.60 · 0%Debug and repair#18Obj. 55.60 · 0%Refactoring#16Obj. 55.60 · 0%Code review#16Obj. 57.70 · 0%Large-repo agent work#35Obj. 30.40 · 0%Frontend / UI#16Obj. 55.60 · 0%

Evidence state

Signal coverage

Benchmarks: 2
Rank points: 7
Usage: 2
Agent: 0
Sources: 7

Objective evidence

Benchmarks and public signals

Swebench: —
Livecodebench: —
Aider: —
Artificial Analysis: 74.3
Openrouter Usage: 1.9
Kilo Usage: —

Persistent history

Recent ranking points

Aug 1#28 · 38

Aug 1#28 · 38

Jul 31#29 · 38

Jul 31#29 · 38

Jul 31#29 · 38

Jul 31#29 · 38

Usage signals

OpenRouter1.9%

Overall1.3%

Agent and tool signals

Evidence is being collected. Missing data is not treated as zero.

Source trail

Anthropic recommends Opus 4.8 for complex agentic coding and enterprise work.Artificial Analysis Coding Agents · ok WebDev Arena · ok Claude Platform models · ok Kilo leaderboard · ok GLM-5 repository · ok Kimi K2.7 Code model card · ok LongCat 2.0 model card · ok