Benchmark Almanac

Nous Research

We track 4 benchmarks and 1 arena that Nous Research made or helped make. Its models are on its lab page.

0 in use 1 with public code 3 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
RefusalBench (Nous Research) Models get requests in scenarios that closed and open models commonly refuse. The score is the percent of questions the model answers, averaged over 5 trials. Deception and alignment Aug 2025 57.1% Fading
Hermes Bench Agents in Hermes Agent do 150 tasks across 25 categories, such as research, diagrams, memory, tool use and safety. Graders check the files and state left behind. General assistants Oct 2026 76.7% Not in cards
Hermes Index Nous Research averages pass@1 and cost per task over Hermes Bench, Terminal-Bench 4.0, Terminal-Bench Science and SkillsBench, all run in Hermes Agent. Composite indices Oct 2026 63.3 Not in cards
Censorship & Propaganda Eval Models answer 803 prompts about censored topics and official narratives. A judge scores factual coverage and endorsement of propaganda separately. Bias, fairness and politics Nov 2026 Not in cards

Arenas

Name Kind Released Best score Status
Husky Hold'em Bench Models write Python poker bots and revise them after practice games. Six bots compete at each table, and the tournament tracks average winnings. Multi-agent games Jul 2025 Not in cards