AI & ML interests
We are dedicated to advancing the field of natural language processing, in collaboration with the open-source community, through bleeding-edge research and a commitment to symbiotic development.
Papers
Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation
Targeted Neuron Modulation via Contrastive Pair Search
The Hermes 3 Series of Models
-
Hermes 3 Technical Report
Paper • 2408.11857 • Published • 60 -
NousResearch/Hermes-3-Llama-3.1-405B
Text Generation • 406B • Updated • 561 • 273 -
NousResearch/Hermes-3-Llama-3.1-70B
Text Generation • 71B • Updated • 4.61k • • 135 -
NousResearch/Hermes-3-Llama-3.1-8B
Text Generation • 8B • Updated • 420k • • 492
Nous' Flagship LLM Series
-
NousResearch/Hermes-2-Theta-Llama-3-70B
Text Generation • 71B • Updated • 2.36k • • 83 -
NousResearch/Hermes-2-Pro-Llama-3-70B
Text Generation • 71B • Updated • 181 • • 35 -
NousResearch/Hermes-2-Theta-Llama-3-8B
Text Generation • 8B • Updated • 11k • • 208 -
NousResearch/Hermes-2-Pro-Llama-3-8B
Text Generation • 8B • Updated • 4.68k • • 456
allenai's Dolma dataset as native Hugging Face datasets
Evals from the Hermes-4 Technical Report
-
NousResearch/eval-Hermes-4-405B-reasoning
Viewer • Updated • 103k • 339 • 8 -
NousResearch/eval-Hermes-4-405B-nonreasoning
Viewer • Updated • 102k • 164 • 4 -
NousResearch/eval-Hermes-4-70B-reasoning
Viewer • Updated • 103k • 199 • 5 -
NousResearch/eval-Hermes-4-14B-nonreasoning-old
Viewer • Updated • 103k • 152 • 3
Preview models of hybrid reasoner Hermes series
-
NousResearch/DeepHermes-3-Mistral-24B-Preview
Text Generation • 24B • Updated • 537 • • 124 -
NousResearch/DeepHermes-3-Llama-3-8B-Preview
Text Generation • 8B • Updated • 1.29k • • 369 -
NousResearch/DeepHermes-3-Llama-3-3B-Preview
Text Generation • 3B • Updated • 768 • • 39 -
NousResearch/DeepHermes-3-Mistral-24B-Preview-GGUF
24B • Updated • 384 • 37
A collection of experimental artifacts created with Atropos, Nous' RL Environments framework - https://github.com/NousResearch/Atropos
-
NousResearch/SWE-smith-oracle
Viewer • Updated • 10.2k • 78 • 5 -
NousResearch/DeepHermes-ToolCalling-Specialist-Atropos
Reinforcement Learning • 8B • Updated • 87 • 20 -
NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos
Text Generation • 8B • Updated • 129 • • 16 -
NousResearch/DeepHermes-Egregore-v2-RLAIF-8b-Atropos
Reinforcement Learning • 8B • Updated • 76 • 7
Extension of Llama 2 to 128k context windows
-
YaRN: Efficient Context Window Extension of Large Language Models
Paper • 2309.00071 • Published • 86 -
NousResearch/Yarn-Mistral-7b-128k
Text Generation • Updated • 782 • 574 -
NousResearch/Yarn-Solar-10b-64k
Text Generation • Updated • 102 • 16 -
NousResearch/Yarn-Llama-2-70b-32k
Text Generation • Updated • 98 • 38
Evals from the Hermes-4 Technical Report
-
NousResearch/eval-Hermes-4-405B-reasoning
Viewer • Updated • 103k • 339 • 8 -
NousResearch/eval-Hermes-4-405B-nonreasoning
Viewer • Updated • 102k • 164 • 4 -
NousResearch/eval-Hermes-4-70B-reasoning
Viewer • Updated • 103k • 199 • 5 -
NousResearch/eval-Hermes-4-14B-nonreasoning-old
Viewer • Updated • 103k • 152 • 3
Preview models of hybrid reasoner Hermes series
-
NousResearch/DeepHermes-3-Mistral-24B-Preview
Text Generation • 24B • Updated • 537 • • 124 -
NousResearch/DeepHermes-3-Llama-3-8B-Preview
Text Generation • 8B • Updated • 1.29k • • 369 -
NousResearch/DeepHermes-3-Llama-3-3B-Preview
Text Generation • 3B • Updated • 768 • • 39 -
NousResearch/DeepHermes-3-Mistral-24B-Preview-GGUF
24B • Updated • 384 • 37
The Hermes 3 Series of Models
-
Hermes 3 Technical Report
Paper • 2408.11857 • Published • 60 -
NousResearch/Hermes-3-Llama-3.1-405B
Text Generation • 406B • Updated • 561 • 273 -
NousResearch/Hermes-3-Llama-3.1-70B
Text Generation • 71B • Updated • 4.61k • • 135 -
NousResearch/Hermes-3-Llama-3.1-8B
Text Generation • 8B • Updated • 420k • • 492
A collection of experimental artifacts created with Atropos, Nous' RL Environments framework - https://github.com/NousResearch/Atropos
-
NousResearch/SWE-smith-oracle
Viewer • Updated • 10.2k • 78 • 5 -
NousResearch/DeepHermes-ToolCalling-Specialist-Atropos
Reinforcement Learning • 8B • Updated • 87 • 20 -
NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos
Text Generation • 8B • Updated • 129 • • 16 -
NousResearch/DeepHermes-Egregore-v2-RLAIF-8b-Atropos
Reinforcement Learning • 8B • Updated • 76 • 7
Nous' Flagship LLM Series
-
NousResearch/Hermes-2-Theta-Llama-3-70B
Text Generation • 71B • Updated • 2.36k • • 83 -
NousResearch/Hermes-2-Pro-Llama-3-70B
Text Generation • 71B • Updated • 181 • • 35 -
NousResearch/Hermes-2-Theta-Llama-3-8B
Text Generation • 8B • Updated • 11k • • 208 -
NousResearch/Hermes-2-Pro-Llama-3-8B
Text Generation • 8B • Updated • 4.68k • • 456
Extension of Llama 2 to 128k context windows
-
YaRN: Efficient Context Window Extension of Large Language Models
Paper • 2309.00071 • Published • 86 -
NousResearch/Yarn-Mistral-7b-128k
Text Generation • Updated • 782 • 574 -
NousResearch/Yarn-Solar-10b-64k
Text Generation • Updated • 102 • 16 -
NousResearch/Yarn-Llama-2-70b-32k
Text Generation • Updated • 98 • 38
allenai's Dolma dataset as native Hugging Face datasets