stack-nousresearch.stack.json json {
"osr": "1",
"kind": "stack",
"name": "Nous Research tiers",
"description": "Nous Research only: Hermes 3 70B Instruct for everyday requests, Hermes 3 405B Instruct for moderate work, Hermes 4 405B for hard ones - with complexity rules and a cost-aware objective.",
"targets": [
{
"ocm": "1",
"id": "model-nousresearch-hermes-3-llama-3-1-70b",
"kind": "llm",
"name": "Hermes 3 70B Instruct",
"description": "Hermes 3 is a generalist language model with many improvements over [Hermes 2](/models/nousresearch/nous-hermes-2-mistral-7b-dpo), including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coherence, and improvements across the...",
"publisher": "Nous Research",
"version": "1.0.0",
"capabilities": {
"domains": [
"general"
],
"tags": [
"llm",
"nousresearch",
"openrouter",
"open-weights",
"models"
],
"languages": [
"en"
],
"modalities": [
"text"
],
"supports_tools": false,
"supports_streaming": true,
"context_window": 131072
},
"quality_prior": 0.6,
"examples": [
"Hermes 3 is a generalist language model with many improvements over [Hermes 2](/models/nousresearch/nous-hermes-2-mistral-7b-dpo), including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coherence, and improvements across the..."
],
"primary": false,
"metadata": {
"source": {
"provider": "models",
"ref": "1723939200",
"url": "https://openrouter.ai/nousresearch/hermes-3-llama-3.1-70b",
"key": "nousresearch/hermes-3-llama-3.1-70b",
"catalogue": "https://openrouter.ai/api/v1"
},
"model": "nousresearch/hermes-3-llama-3.1-70b",
"pricing": {
"input_usd_per_1m": 0.7,
"output_usd_per_1m": 0.7
},
"max_output_tokens": 16384,
"output_modalities": [
"text"
],
"hugging_face_id": "NousResearch/Hermes-3-Llama-3.1-70B",
"created": 1723939200,
"benchmarks": {
"ifeval": 76.6,
"bbh": 67.6,
"math_lvl5": 21,
"gpqa": 36.2,
"musr": 49.5,
"mmlu_pro": 47.3,
"hf_leaderboard_avg": 49.7,
"hub_downloads": 3142,
"hub_downloads_total": 331480,
"hub_likes": 138,
"hub_trending": 0,
"osr_index": 35.8,
"osr_coverage": 0.3
}
},
"cost": {
"usd_per_1k_tokens": 0.0007
},
"endpoints": [
{
"protocol": "openai",
"url": "https://openrouter.ai/api/v1"
}
]
},
{
"ocm": "1",
"id": "model-nousresearch-hermes-3-llama-3-1-405b",
"kind": "llm",
"name": "Hermes 3 405B Instruct",
"description": "Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coherence, and improvements across the...",
"publisher": "Nous Research",
"version": "1.0.0",
"capabilities": {
"domains": [
"general"
],
"tags": [
"llm",
"nousresearch",
"openrouter",
"open-weights",
"models"
],
"languages": [
"en"
],
"modalities": [
"text"
],
"supports_tools": false,
"supports_streaming": true,
"context_window": 131072
},
"quality_prior": 0.6,
"examples": [
"Hermes 3 is a generalist language model with many improvements over Hermes 2, including advanced agentic capabilities, much better roleplaying, reasoning, multi-turn conversation, long context coherence, and improvements across the..."
],
"primary": false,
"metadata": {
"source": {
"provider": "models",
"ref": "1723766400",
"url": "https://openrouter.ai/nousresearch/hermes-3-llama-3.1-405b",
"key": "nousresearch/hermes-3-llama-3.1-405b",
"catalogue": "https://openrouter.ai/api/v1"
},
"model": "nousresearch/hermes-3-llama-3.1-405b",
"pricing": {
"input_usd_per_1m": 1,
"output_usd_per_1m": 1
},
"max_output_tokens": 16384,
"output_modalities": [
"text"
],
"hugging_face_id": "NousResearch/Hermes-3-Llama-3.1-405B",
"created": 1723766400,
"benchmarks": {
"hub_downloads": 1227,
"hub_downloads_total": 956645,
"hub_likes": 274,
"hub_trending": 0
}
},
"cost": {
"usd_per_1k_tokens": 0.001
},
"endpoints": [
{
"protocol": "openai",
"url": "https://openrouter.ai/api/v1"
}
]
},
{
"ocm": "1",
"id": "model-nousresearch-hermes-4-405b",
"kind": "llm",
"name": "Hermes 4 405B",
"description": "Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research. It introduces a hybrid reasoning mode, where the model can choose to deliberate internally with...",
"publisher": "Nous Research",
"version": "1.0.0",
"capabilities": {
"domains": [
"math"
],
"tags": [
"llm",
"nousresearch",
"openrouter",
"open-weights",
"reasoning",
"models"
],
"languages": [
"en"
],
"modalities": [
"text"
],
"supports_tools": false,
"supports_streaming": true,
"context_window": 131072
},
"quality_prior": 0.6,
"examples": [
"Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research. It introduces a hybrid reasoning mode, where the model can choose to deliberate internally with..."
],
"primary": false,
"metadata": {
"source": {
"provider": "models",
"ref": "1756235463",
"url": "https://openrouter.ai/nousresearch/hermes-4-405b",
"key": "nousresearch/hermes-4-405b",
"catalogue": "https://openrouter.ai/api/v1"
},
"model": "nousresearch/hermes-4-405b",
"pricing": {
"input_usd_per_1m": 1,
"output_usd_per_1m": 3
},
"max_output_tokens": 117964,
"output_modalities": [
"text"
],
"hugging_face_id": "NousResearch/Hermes-4-405B",
"reasoning": true,
"created": 1756235463,
"benchmarks": {
"hub_downloads": 758,
"hub_downloads_total": 6504,
"hub_likes": 95,
"hub_trending": 1
}
},
"cost": {
"usd_per_1k_tokens": 0.002
},
"endpoints": [
{
"protocol": "openai",
"url": "https://openrouter.ai/api/v1"
}
]
}
],
"rules": [
{
"name": "hard-to-strongest",
"when": {
"min_complexity": 0.7
},
"prefer": [
"model-nousresearch-hermes-4-405b"
]
},
{
"name": "easy-stays-cheap",
"when": {
"max_complexity": 0.3
},
"avoid": [
"model-nousresearch-hermes-4-405b"
]
}
],
"objective": {
"quality": 0.7,
"cost": 0.8,
"latency": 0.2
},
"metadata": {
"source": {
"provider": "catalogue-stacks",
"path": "nousresearch",
"ref": "2026-09-19",
"url": "https://openrouter.ai/nousresearch",
"key": "catalogue-stacks/nousresearch",
"catalogue": "https://openrouter.ai/api/v1",
"generated": true
}
}
}
