Nous Research — AI model benchmarks
AI models from Nous Research. 22 models tracked, 22 with a unified ELO rating. Current best: Cogito v2.1 (Reasoning) (ELO 1697, rank #274 overall). Most recent tracked release: 2025-09-01.
Rated models
| Rank | Model | Provider | Unified ELO | Benchmarks |
|---|---|---|---|---|
| 274 | Cogito v2.1 (Reasoning) | Nous Research | 1697 ± 28 | 18 |
| 353 | Hermes 4 405B (Thinking) | Nous Research | 1656 ± 33 | 21 |
| 373 | Hermes 4 405B | Nous Research | 1648 ± 22 | 78 |
| 437 | Hermes 4 70B (Thinking) | Nous Research | 1619 ± 35 | 20 |
| 541 | Hermes 4 405B (Non-reasoning) | Nous Research | 1577 ± 37 | 19 |
| 547 | Hermes 4 70B | Nous Research | 1575 ± 15 | 93 |
| 576 | Hermes-3-Llama-3.1-70B-lorablated | Nous Research | 1566 ± 44 | 9 |
| 695 | Hermes-4-14B | Nous Research | 1527 ± 10 | 37 |
| 716 | DeepHermes-3-Mistral-24B-Preview | Nous Research | 1522 ± 40 | 7 |
| 790 | Hermes 4 70B (Non-reasoning) | Nous Research | 1502 ± 39 | 19 |
| 901 | Nous-Hermes-2-Yi-34B | Nous Research | 1472 ± 19 | 8 |
| 1021 | Hermes-3-Llama-3.1-8B | Nous Research | 1445 ± 21 | 24 |
| 1076 | Nous-Hermes-2-Mixtral-8x7B-DPO | Nous Research | 1433 ± 15 | 23 |
| 1141 | Nous-Hermes-2-Mixtral-8x7B-SFT | Nous Research | 1418 ± 46 | 7 |
| 1161 | Hermes-2-Theta-Llama-3-8B | Nous Research | 1414 ± 16 | 17 |
Interactive version: theaggregate.ai/company?slug=nous-research · How the rankings work · Data refreshed daily, snapshot 2026-07-22.