Which is better, Claude Opus 4.8 or HyperNova 60B 2605 (high, based on gpt-oss-120b)?

Refresh Models —Deals —Regulatory —Sources —

Claude Opus 4.8 vs HyperNova 60B 2605 (high, based on gpt-oss-120b)

17 SHARED BENCHMARKS

Across 17 shared benchmarks, Claude Opus 4.8 scores higher on 16 and HyperNova 60B 2605 (high, based on gpt-oss-120b) on 1. The widest gap is AA Agentic Index, where Claude Opus 4.8 scores 49.4 against 6.8.

ANTHROPICVSOPENAI17 SHARED16–1 HEAD-TO-HEAD

BenchmarkClaude Opus 4.8MARGINHyperNova 60B 2605 (high, based on gpt-oss-120b)

AA Agentic Index49.46.8

AA Intelligence57.318.3

AA-LCR7335.3

AA-Omniscience28.8-54.6

Artificial Analysis Coding Index74.323.2

critpt20.90.9

gdpval53.98

GPQA Diamond93.673.3

HLE57.916.6

IFBench62.266.5

OmniScience Accuracy48.815.2

OmniScience Non-Hallucination60.717.7

scicode53.533

TauBench V3 - Banking34.25.6

Terminal-Bench 2.18518.4

Terminal-Bench Hard58.323.5

τ²-Bench Telecom (AA run)94.463.2

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.