VECTOR WIREAI INTELLIGENCE
UTC
Refresh Models Deals Regulatory Sources

Claude Fable 5.1 vs Claude Opus 5

49 SHARED BENCHMARKS

Across 49 shared benchmarks, Claude Fable 5.1 scores higher on 37 and Claude Opus 5 on 9, with 3 level. The widest gap is Terminal-Bench-Science 0.1, where Claude Fable 5.1 scores 52.6 against 29. Claude Opus 5 is the cheaper of the two on tracked API pricing ($5.00 against $10.00 per million input tokens).

ANTHROPICVSANTHROPIC49 SHARED379 HEAD-TO-HEAD
AA-LCR8078.7
arc_agi_197.597.5
ARC-AGI-29090.4
critpt31.129.1
DeepSWE 1.167.468.8
gdpval67.766.2
GDPval-AA v218531861
GMMLU9492.5
GPQA Diamond93.793.7
HLE6564.7
MILU9392.1
officeqa80.278.1
OSWorld 2.077.970.6
OSWorld 2.0 (partial)77.975.4
OSWorld 2.0 (strict)41.739.6
scicode6255.7
Terminal-Bench 4.05652
Terminal-Bench-Science 0.152.629
Toolathlon Verified Pass³73.173.1

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, anthropic-official), otherwise the lowest tracked offer.

Compare Claude Opus 5 withEVERY TRACKED PAIRING

Anthropic11 PAIRINGS
OpenAI11 PAIRINGS
Google11 PAIRINGS
Alibaba11 PAIRINGS
DeepSeek5 PAIRINGS
Moonshot4 PAIRINGS
Meta2 PAIRINGS
Z.ai2 PAIRINGS
MiniMax1 PAIRING
NVIDIA1 PAIRING