Model creator
Nous Research
Every Nous Research model in the BenchSift catalogue, with its scores, ranks, prices and capabilities as the sources publish them.
- Models
- 8
- 10 configurations, reasoning levels included
- Open weights
- 8
- 8 of 8 models
- Latest release
- Aug 27, 2025
- Hermes 4 - Llama-3.1 405B (Non-reasoning) · First dated model: Apr 30, 2024
- Median price
- $1.25
- $0.70 to $1.50 per million tokens, across 4 priced models
Models
8Reasoning levels are folded into one entry. The rank is the whole catalogue's: sorting does not change it.
- Nous: Hermes 4 405B7.5#557 of 782$1.50—
Text · Reasoning · Open weights · Aug 26, 2025 · $1.50
- Hermes 4 - Llama-3.1 405B7.4#567 of 782$1.5031.9 t/s
Text · Reasoning · Open weights · 2 configurations · Aug 27, 2025 · $1.50 · 31.9 t/s
- Nous: Hermes 3 405B Instruct7.4#567 of 782$1—
Text · Open weights · Aug 16, 2024 · $1
- Hermes 4 - Llama-3.1 70B6.7#612 of 782——
Text · Reasoning · Open weights · 2 configurations · Aug 27, 2025
- DeepHermes 3 - Mistral 24B Preview (Non-reasoning)6.1#657 of 782——
Text · Open weights · Mar 13, 2025
- Hermes 3 - Llama-3.1 70B6.0#660 of 782$0.70176 t/s
Text · Open weights · Aug 15, 2024 · $0.70 · 176 t/s
- DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)5.1#739 of 782——
Text · Open weights · Feb 13, 2025
- Hermes 2 Pro Llama 3 8B———
Text · Open weights · Apr 30, 2024
Where these figures come from
Each catalogue model is tied to its creator by the identity Artificial Analysis publishes, or OpenRouter's and Models.dev's for models it does not track. Counts fold the reasoning levels of one model together. Ranks compare each measure with every measured model in the catalogue, configurations included, as model pages do. The price is the published blended price (3 input tokens to 1 output token).
Figures computed by BenchSift from the catalogue: Artificial Analysis, OpenRouter, Models.dev and Hugging Face. A value no model carries is not shown.
Data retrieved on Nov 12, 2026.