stack-open-weights.stack.json json Copy{
"osr" : "1" ,
"kind" : "stack" ,
"name" : "Open-weights models" ,
"description" : "Hosted open-weights models only - the same stack runs on your own hardware later." ,
"targets" : [
{
"ocm" : "1" ,
"id" : "model-cohere-command-r-plus-08-2024" ,
"kind" : "llm" ,
"name" : "Command R+ (08-2024)" ,
"description" : "command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint..." ,
"publisher" : "Cohere" ,
"version" : "1.0.0" ,
"capabilities" : {
"domains" : [
"coding" ,
"math"
],
"tags" : [
"llm" ,
"cohere" ,
"openrouter" ,
"open-weights" ,
"tool-calling" ,
"models"
],
"languages" : [
"en"
],
"modalities" : [
"text"
],
"supports_tools" : true ,
"supports_streaming" : true ,
"context_window" : 128000
},
"quality_prior" : 0.6 ,
"examples" : [
"command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint..."
],
"primary" : false ,
"metadata" : {
"source" : {
"provider" : "models" ,
"ref" : "1724976000" ,
"url" : "https://openrouter.ai/cohere/command-r-plus-08-2024" ,
"key" : "cohere/command-r-plus-08-2024" ,
"catalogue" : "https://openrouter.ai/api/v1"
},
"model" : "cohere/command-r-plus-08-2024" ,
"pricing" : {
"input_usd_per_1m" : 2.5 ,
"output_usd_per_1m" : 10
},
"max_output_tokens" : 4000 ,
"output_modalities" : [
"text"
],
"hugging_face_id" : "CohereLabs/c4ai-command-r-plus-08-2024" ,
"created" : 1724976000 ,
"benchmarks" : {
"hub_downloads" : 61 ,
"hub_downloads_total" : 262549 ,
"hub_likes" : 306 ,
"hub_trending" : 0
}
},
"cost" : {
"usd_per_1k_tokens" : 0.00625
},
"endpoints" : [
{
"protocol" : "openai" ,
"url" : "https://openrouter.ai/api/v1"
}
]
},
{
"ocm" : "1" ,
"id" : "model-moonshotai-kimi-k3-1-pro" ,
"kind" : "llm" ,
"name" : "Kimi K3.1 Pro" ,
"description" : "Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D..." ,
"publisher" : "Moonshot AI" ,
"version" : "1.0.0" ,
"capabilities" : {
"domains" : [
"coding" ,
"math"
],
"tags" : [
"llm" ,
"moonshotai" ,
"openrouter" ,
"open-weights" ,
"reasoning" ,
"tool-calling" ,
"image" ,
"video" ,
"models"
],
"languages" : [
"en"
],
"modalities" : [
"text" ,
"image" ,
"video"
],
"supports_tools" : true ,
"supports_streaming" : true ,
"context_window" : 1048576
},
"quality_prior" : 0.465 ,
"examples" : [
"Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D..."
],
"primary" : false ,
"metadata" : {
"source" : {
"provider" : "models" ,
"ref" : "1792778188" ,
"url" : "https://openrouter.ai/moonshotai/kimi-k3.1-pro" ,
"key" : "moonshotai/kimi-k3.1-pro" ,
"catalogue" : "https://openrouter.ai/api/v1"
},
"model" : "moonshotai/kimi-k3.1-pro" ,
"pricing" : {
"input_usd_per_1m" : 0.253 ,
"output_usd_per_1m" : 8.8 ,
"cache_read_usd_per_1m" : 0.18
},
"max_output_tokens" : 943718 ,
"output_modalities" : [
"text"
],
"hugging_face_id" : "moonshotai/Kimi-K3.1-Pro" ,
"reasoning" : true ,
"created" : 1792778188 ,
"benchmarks" : {
"intelligence_index" : 46.5 ,
"hub_downloads" : 68001 ,
"hub_downloads_total" : 68001 ,
"hub_likes" : 2342 ,
"hub_trending" : 53.5 ,
"osr_index" : 47.3 ,
"osr_coverage" : 0.4
}
},
"cost" : {
"usd_per_1k_tokens" : 0.004527
},
"endpoints" : [
{
"protocol" : "openai" ,
"url" : "https://openrouter.ai/api/v1"
}
]
},
{
"ocm" : "1" ,
"id" : "model-qwen-qwen3-8-2-4t-a95b" ,
"kind" : "llm" ,
"name" : "Qwen3.8 2.4T A95B" ,
"description" : "Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of [Qwen3.8 Max](/qwen/qwen3.8-max), with 95 billion active parameters out of 2.4 trillion total. It is..." ,
"publisher" : "Qwen" ,
"version" : "1.0.0" ,
"capabilities" : {
"domains" : [
"coding" ,
"math"
],
"tags" : [
"llm" ,
"qwen" ,
"openrouter" ,
"open-weights" ,
"reasoning" ,
"tool-calling" ,
"models"
],
"languages" : [
"en"
],
"modalities" : [
"text"
],
"supports_tools" : true ,
"supports_streaming" : true ,
"context_window" : 1048576
},
"quality_prior" : 0.399 ,
"examples" : [
"Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of [Qwen3.8 Max](/qwen/qwen3.8-max), with 95 billion active parameters out of 2.4 trillion total. It is..."
],
"primary" : false ,
"metadata" : {
"source" : {
"provider" : "models" ,
"ref" : "1786551702" ,
"url" : "https://openrouter.ai/qwen/qwen3.8-2.4t-a95b" ,
"key" : "qwen/qwen3.8-2.4t-a95b" ,
"catalogue" : "https://openrouter.ai/api/v1"
},
"model" : "qwen/qwen3.8-2.4t-a95b" ,
"pricing" : {
"input_usd_per_1m" : 2 ,
"output_usd_per_1m" : 6 ,
"cache_read_usd_per_1m" : 0.25
},
"max_output_tokens" : 65536 ,
"output_modalities" : [
"text"
],
"hugging_face_id" : "Qwen/Qwen3.8-2.4T-A95B" ,
"reasoning" : true ,
"created" : 1786551702 ,
"benchmarks" : {
"coding_index" : 71.9 ,
"agentic_index" : 50.1 ,
"intelligence_index" : 39.9 ,
"hub_downloads" : 29851 ,
"hub_likes" : 1407 ,
"hub_downloads_total" : 131282 ,
"hub_trending" : 13 ,
"osr_index" : 43.5 ,
"osr_coverage" : 0.4
}
},
"cost" : {
"usd_per_1k_tokens" : 0.004
},
"endpoints" : [
{
"protocol" : "openai" ,
"url" : "https://openrouter.ai/api/v1"
}
]
},
{
"ocm" : "1" ,
"id" : "model-anthracite-org-magnum-v4-72b" ,
"kind" : "llm" ,
"name" : "Magnum v4 72B" ,
"description" : "This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus). The model is fine-tuned on top of [Qwen2.5 72B](https://openrouter.ai/qwen/qwen-2.5-72b-instruct)." ,
"publisher" : "Anthracite Org" ,
"version" : "1.0.0" ,
"capabilities" : {
"domains" : [
"coding" ,
"math"
],
"tags" : [
"llm" ,
"anthracite-org" ,
"openrouter" ,
"open-weights" ,
"models"
],
"languages" : [
"en"
],
"modalities" : [
"text"
],
"supports_tools" : false ,
"supports_streaming" : true ,
"context_window" : 32768
},
"quality_prior" : 0.6 ,
"examples" : [
"This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus). The model is fine-tuned on top of [Qwen2.5 72B](https://openrouter.ai/qwen/qwen-2.5-72b-instruct)."
],
"primary" : false ,
"metadata" : {
"source" : {
"provider" : "models" ,
"ref" : "1729555200" ,
"url" : "https://openrouter.ai/anthracite-org/magnum-v4-72b" ,
"key" : "anthracite-org/magnum-v4-72b" ,
"catalogue" : "https://openrouter.ai/api/v1"
},
"model" : "anthracite-org/magnum-v4-72b" ,
"pricing" : {
"input_usd_per_1m" : 2.5 ,
"output_usd_per_1m" : 5 ,
"cache_read_usd_per_1m" : 2.5
},
"max_output_tokens" : 4096 ,
"output_modalities" : [
"text"
],
"hugging_face_id" : "anthracite-org/magnum-v4-72b" ,
"created" : 1729555200 ,
"benchmarks" : {
"hub_downloads" : 1625 ,
"hub_likes" : 66 ,
"hub_downloads_total" : 51193 ,
"hub_trending" : 2
}
},
"cost" : {
"usd_per_1k_tokens" : 0.00375
},
"endpoints" : [
{
"protocol" : "openai" ,
"url" : "https://openrouter.ai/api/v1"
}
]
}
],
"rules" : [
{
"name" : "hard-to-strongest" ,
"when" : {
"min_complexity" : 0.7
},
"prefer" : [
"model-cohere-command-r-plus-08-2024"
]
}
],
"objective" : {
"quality" : 0.7 ,
"cost" : 0.8 ,
"latency" : 0.2
},
"metadata" : {
"source" : {
"provider" : "catalogue-stacks" ,
"path" : "open-weights" ,
"ref" : "2026-11-22" ,
"url" : "https://openrouter.ai/models" ,
"key" : "catalogue-stacks/open-weights" ,
"catalogue" : "https://openrouter.ai/api/v1" ,
"generated" : true
}
}
}