Which is better, Claude Fable 5 or MiniCPM5-2B?

Refresh Models —Deals —Regulatory —Sources —

Claude Fable 5 vs MiniCPM5-2B

15 SHARED BENCHMARKS

Across 15 shared benchmarks, Claude Fable 5 scores higher on 14 and MiniCPM5-2B on 1. The widest gap is OmniScience Accuracy, where Claude Fable 5 scores 65.3 against 8.4.

ANTHROPICVSOPENBMB15 SHARED14–1 HEAD-TO-HEAD

BenchmarkClaude Fable 5MARGINMiniCPM5-2B

AA Agentic Index519

AA Intelligence6213.1

AA-LCR82.359.3

AA-Omniscience43.3-11.7

Artificial Analysis Coding Index76.514.5

critpt28.60.3

gdpval56.616.4

GPQA Diamond92.670.2

HLE64.58.9

OmniScience Accuracy65.38.4

OmniScience Non-Hallucination36.478.1

scicode6126.3

TauBench V3 - Banking38.120.8

Terminal-Bench 2.1888.6

terminalbenchV4042.40

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.