Releases - NOUS RESEARCH

NOUS RELEASES

	PROJECT NAME	TYPE	DETAILS	RELEASE DATE	SIZE
0	Hermes Agent	AGENT	An autonomous agent that lives on your server, remembers what it learns, and gets more capable the longer it runs.	02/25/26
1	NousCoder-14B	MODEL	Competitive olympiad programming post-trained on Qwen-3-14B	01/6/26	29.6
2	Nomos 1	MODEL	Small but mighty 30B SOTA mathematician in partnership with Hillclimb	12/9/25	120
3	Hermes-4.3-Seed-36B	MODEL	Roughly equivalent performance to Hermes-4-70B at half the model size; post-trained entirely on the Psyche network	12/3/25	72
4	Hermes 4 Technical Report	PAPER	Technical Report	08/26/25
5	Hermes‑4‑Llama‑3.1‑405B	MODEL	Frontier hybrid‑mode reasoning model based on Llama‑3.1‑405B	08/26/25	802
6	Hermes‑4‑Llama‑3.1‑70B	MODEL	Smaller hybrid‑mode reasoning model with 70B parameters. Shares the same improvements as the 405B variant	08/26/25	140
7	Hermes‑4‑14B	MODEL	Small and dense Hermes variant for local inference.	08/26/25	28
8	Measuring Thinking Efficiency in Reasoning Models	RESEARCH	Measuring Thinking Efficiency in Reasoning Models: The Missing Benchmark	08/14/25
9	Hermes 3 Dataset	DATASET	Complete dataset used in pretraining of Hermes 3 models	07/11/25
10	Hermes Agent	RESEARCH	Taming LLMs with Sequential Monte Carlo	06/1/25
11	Psyche Network	TRAINING	Open infrastructure democratizing AI development	05/14/25
12	Atropos	FRAMEWORK	Language Model RL Environments	04/29/25
13	DeepHermes-3-Llama-3-8B-Preview	MODEL	Model with deep reasoning and chat modes	02/14/25	16.1
14	Hermes-3-Llama-3.2-3B	MODEL	Model that can run on most modern hardware	12/11/24	6.5
15	GOD	SIMULATOR	Social agent	11/29/24
16	S8N	SIMULATOR	Social agent	11/29/24
17	DeMo: Decoupled Momentum Optimization	PAPER	Technical report	11/29/24
18	Hermes Chat	SIMULATOR	Chat Interface	11/8/24
19	TEE	SIMULATOR	Autonomous agent	10/29/24
20	A Preliminary Report On DisTrO	PAPER	Technical report	08/26/24
21	Hermes 3 405B	MODEL	Model	08/24/24	812
22	Original Nous WorldSim Prompt	PROMPT	Prompt engineering	03/15/24
23	Hermes 3 Technical Report	PAPER	Technical report	08/14/24
24	Hermes 2 Pro 70B	MODEL	Function-calling model	06/27/24	42.5
25	Hermes 2 Theta	MODEL	Model	06/20/24	42.5
26	Character Codex	DATASET	Roleplaying dataset	06/11/24	0.0112
27	Nous WorldSim	CODE	CLI simulator	03/23/24
28	Hermes-2-Pro-Mistral-7B	MODEL	Master function-calling model	03/11/24	14.49
29	NousResearch/Genstruct-7B	MODEL	Data synthesis model	03/7/24	15
30	Nous Research Official Branding Booklet	PAPER	Design & story of Nous	03/5/24	0.042
31	NousResearch/Nous-Hermes-2-Mistral-7B-DPO	MODEL	7B model	02/20/24	15
32	NousResearch/finetuning-subnet	SUBNET	Evals on-chain	01/29/24	0.0001
33	NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO	MODEL	MoE model + DPO	01/15/24	95
34	NousResearch/Nous-Hermes-2-Mixtral-8x7B-SFT	MODEL	MoE model - DPO	01/15/24	95
35	NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO-GGUF	MODEL	Quantized MoE model	01/15/24	17-50
36	NousResearch/Nous-Hermes-2-Yi-34B	MODEL	Hermes on Yi	12/25/23	70
37	NousResearch/Nous-Capybara-34B	MODEL	Capybara-series model	11/14/23	70
38	NousResearch/Yarn-Mistral-7b-128k	MODEL	Long-context model	11/2/23	15
39	NousResearch/Obsidian-3B-V0.5	MODEL	Multimodal vision LLM	10/30/23	5
40	NousResearch/Obsidian	CODE	Code for obsidian model	10/30/23	0.015
41	YaRN: Efficient Context Window Extension of Large Language Models	PAPER	Architecture behind YaRN models	08/31/23	0.0007
42	NousResearch/Nous-Hermes-Llama2-13b	MODEL	Hermes on Llama2	07/21/23	25