Claude Opus 4.1 vs Claude Fable 5 vs Claude Opus 4.8

Best cell per row highlighted. Null means undisclosed — never counted as zero.

Anthropic
Claude Opus 4.1GA
5.5
AI Panel
Anthropic
Claude Fable 5GA
9.5
AI Panel
Anthropic
Claude Opus 4.8GA
9.2
AI Panel

Identity & lifecycle

ProviderAnthropicAnthropicAnthropic
Family / tierClaude 4 · OpusClaude 5 · FableClaude 4 · Opus
StatusGAGAGA
Released2025-08-042026-06-082026-05-27
Knowledge cutoff2025-01

Architecture & context

Context window200K1M1M
Max output tokens32K128K128K
Input modalitiestext, imagetext, image, filetext, image
Reasoning modeoptionalalwaysalways
Open weightsNoNoNo

Pricing (per Mtok)

Input$15$10$5
Output$75$50$25
Cached input$1.5$1$0.5
Batch input$7.5$5$2.5
Free tierNoNoNo

Speed

Output speed (tok/s)4560
Time to first token (s)1.80.5
Latency tierslowslowmedium

Trust & deployment

Trains on inputsNoNoNo
LicenseProprietaryProprietaryProprietary
CloudsBedrock, Vertex AI, Azure AI FoundryBedrock, Vertex AI, Azure AI FoundryBedrock, Vertex AI, Azure AI Foundry
ComplianceSOC2, HIPAA, GDPR, ISO27001SOC2, HIPAA, GDPR, ISO27001SOC2, HIPAA, GDPR, ISO27001

AI Panel scoring

Unified score5.59.59.2
Decision Maker59.59
Domain Strategist5.59.59
Finance Lead478.5
Domain Practitioner59.59.5
Power User699
Skeptic57.58
Value score2.56.58

Benchmarks

MMLU88.8
GPQA Diamond80.993.6
SWE-bench Verified74.59588.6
MMMU77.1
Terminal-Bench8874.6
LMArena Elo1425
LMArena Coding Elo1475
Artificial Analysis Index33

Scores link to their sources. Missing cells mean the vendor hasn't disclosed a result — honesty over padding.