Neura Intelligence Index

AI Model Benchmarks

Back to Leaderboard

GPT OSS 120B vs Hermes 3 70B

OpenAI 🇺🇸

47.6

#158—

N

Nous Research 🇺🇸

48.8

#155—

Category Comparison

Benchmark-by-Benchmark

reasoning

Benchmark	GPT OSS 120B	Hermes 3 70B
GPQA Diamond	80.1	66.1
Humanity's Last Exam	14.9	—

agent

Benchmark	GPT OSS 120B	Hermes 3 70B
TAU-Bench Retail	67.8	—

Category Winners

reasoning

N

Hermes 3 70B

48.8

agent

GPT OSS 120B

50.1

GPT OSS 120B vs Hermes 3 70B — Benchmark Comparison | Neura Market