MiscSubjects

Collection

Inception

3 model catalog entries

LinkedInXFacebookRedditEdit
Inception: Mercury 3
InceptionMercury 3
inception/mercury-3
2026-10-16text->textreasoning: high / medium / low / none
Design Arena
1175 Elo · #41
svg · 42.8% wins
Input
$0.10
per 1M tokens
Output
$0.35
per 1M tokens
Cached input
$0.01
per 1M tokens
Context
1M
tokens it reads
Max output
131K
tokens it writes
Mercury 3 is a diffusion-based language model from Inception, and the lab's most capable model to date. It is suited for agentic workflows and real-time applications where latency compounds, such...
inspect — the catalog entry
{
  "id": "inception/mercury-3",
  "canonical_slug": "inception/mercury-3-20261019",
  "hugging_face_id": null,
  "name": "Inception: Mercury 3",
  "created": 1792158337,
  "description": "Mercury 3 is a diffusion-based language model from Inception, and the lab's most capable model to date. It is suited for agentic workflows and real-time applications where latency compounds, such...",
  "context_length": 1000000,
  "architecture": {
    "modality": "text->text",
    "input_modalities": [
      "text"
    ],
    "output_modalities": [
      "text"
    ],
    "tokenizer": "Other",
    "instruct_type": null
  },
  "pricing": {
    "prompt": "0.0000001",
    "completion": "0.00000035",
    "input_cache_read": "0.00000001"
  },
  "top_provider": {
    "context_length": 1000000,
    "max_completion_tokens": 131072,
    "is_moderated": false
  },
  "per_request_limits": null,
  "supported_parameters": [
    "include_reasoning",
    "max_tokens",
    "reasoning",
    "reasoning_effort",
    "response_format",
    "stop",
    "structured_outputs",
    "temperature",
    "tool_choice",
    "tools"
  ],
  "default_parameters": {},
  "supported_voices": null,
  "knowledge_cutoff": null,
  "expiration_date": "2026-11-08",
  "links": {
    "details": "/api/v1/models/inception/mercury-3-20261019/endpoints"
  },
  "endpoints": [
    "/api/v1/chat/completions",
    "/api/v1/completions",
    "/api/v1/messages",
    "/api/v1/responses"
  ],
  "benchmarks": {
    "design_arena": [
      {
        "arena": "models",
        "category": "codecategories",
        "elo": 1229,
        "win_rate": 45.4,
        "rank": 52
      },
      {
        "arena": "models",
        "category": "dataviz",
        "elo": 1206,
        "win_rate": 45,
        "rank": 56
      },
      {
        "arena": "models",
        "category": "gamedev",
        "elo": 1216,
        "win_rate": 44.2,
        "rank": 49
      },
      {
        "arena": "models",
        "category": "svg",
        "elo": 1175,
        "win_rate": 42.8,
        "rank": 41
      },
      {
        "arena": "models",
        "category": "uicomponent",
        "elo": 1232,
        "win_rate": 45.5,
        "rank": 48
      },
      {
        "arena": "models",
        "category": "website",
        "elo": 1231,
        "win_rate": 45.7,
        "rank": 54
      }
    ]
  },
  "reasoning": {
    "mandatory": false,
    "default_enabled": true,
    "supported_efforts": [
      "high",
      "medium",
      "low",
      "none"
    ],
    "default_effort": "medium"
  }
}
openrouter.ai/inception/mercury-3
Inception: Mercury 2.5
InceptionMercury 2.5
inception/mercury-2.5
2026-09-08text->textreasoning: high / medium / low / none
Input
$0.04
per 1M tokens
Output
$0.15
per 1M tokens
Cached input
$0
per 1M tokens
Context
260K
tokens it reads
Max output
66K
tokens it writes
Mercury 2.5 is the fastest reasoning LLM, and the latest diffusion LLM (dLLM) from Inception. Instead of generating tokens sequentially, Mercury 2.5 produces and refines multiple tokens in parallel, achieving...
inspect — the catalog entry
{
  "id": "inception/mercury-2.5",
  "canonical_slug": "inception/mercury-2.5-20260908",
  "hugging_face_id": null,
  "name": "Inception: Mercury 2.5",
  "created": 1788892137,
  "description": "Mercury 2.5 is the fastest reasoning LLM, and the latest diffusion LLM (dLLM) from Inception. Instead of generating tokens sequentially, Mercury 2.5 produces and refines multiple tokens in parallel, achieving...",
  "context_length": 260000,
  "architecture": {
    "modality": "text->text",
    "input_modalities": [
      "text"
    ],
    "output_modalities": [
      "text"
    ],
    "tokenizer": "Other",
    "instruct_type": null
  },
  "pricing": {
    "prompt": "0.00000004",
    "completion": "0.00000015",
    "input_cache_read": "0.000000004"
  },
  "top_provider": {
    "context_length": 260000,
    "max_completion_tokens": 65536,
    "is_moderated": false
  },
  "per_request_limits": null,
  "supported_parameters": [
    "include_reasoning",
    "max_tokens",
    "reasoning",
    "reasoning_effort",
    "response_format",
    "stop",
    "structured_outputs",
    "temperature",
    "tool_choice",
    "tools"
  ],
  "default_parameters": {},
  "supported_voices": null,
  "knowledge_cutoff": null,
  "expiration_date": "2026-11-08",
  "links": {
    "details": "/api/v1/models/inception/mercury-2.5-20260908/endpoints"
  },
  "endpoints": [
    "/api/v1/chat/completions",
    "/api/v1/completions",
    "/api/v1/messages",
    "/api/v1/responses"
  ],
  "reasoning": {
    "mandatory": false,
    "default_enabled": true,
    "supported_efforts": [
      "high",
      "medium",
      "low",
      "none"
    ],
    "default_effort": "medium"
  },
  "benchmarks": null
}
openrouter.ai/inception/mercury-2.5
Inception: Mercury 2
InceptionMercury 2
inception/mercury-2
2026-03-04text->textreasoning: high / medium / low / none
Coding index
31.1
Artificial Analysis
Agentic index
1.6
Artificial Analysis
Design Arena
986 Elo · #74
asciiart · 24.2% wins
Input
$0.25
per 1M tokens
Output
$0.75
per 1M tokens
Cached input
$0.02
per 1M tokens
Context
128K
tokens it reads
Max output
50K
tokens it writes
Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM). Instead of generating tokens sequentially, Mercury 2 produces and refines multiple tokens in parallel, achieving...
inspect — the catalog entry
{
  "id": "inception/mercury-2",
  "canonical_slug": "inception/mercury-2-20260304",
  "hugging_face_id": null,
  "name": "Inception: Mercury 2",
  "created": 1772636275,
  "description": "Mercury 2 is an extremely fast reasoning LLM, and the first reasoning diffusion LLM (dLLM). Instead of generating tokens sequentially, Mercury 2 produces and refines multiple tokens in parallel, achieving...",
  "context_length": 128000,
  "architecture": {
    "modality": "text->text",
    "input_modalities": [
      "text"
    ],
    "output_modalities": [
      "text"
    ],
    "tokenizer": "Other",
    "instruct_type": null
  },
  "pricing": {
    "prompt": "0.00000025",
    "completion": "0.00000075",
    "input_cache_read": "0.000000025"
  },
  "top_provider": {
    "context_length": 128000,
    "max_completion_tokens": 50000,
    "is_moderated": false
  },
  "per_request_limits": null,
  "supported_parameters": [
    "include_reasoning",
    "max_tokens",
    "reasoning",
    "reasoning_effort",
    "response_format",
    "stop",
    "structured_outputs",
    "temperature",
    "tool_choice",
    "tools"
  ],
  "default_parameters": {
    "temperature": 0.75,
    "top_p": null,
    "top_k": null,
    "frequency_penalty": null,
    "presence_penalty": null,
    "repetition_penalty": null
  },
  "supported_voices": null,
  "knowledge_cutoff": null,
  "expiration_date": "2026-11-08",
  "links": {
    "details": "/api/v1/models/inception/mercury-2-20260304/endpoints"
  },
  "endpoints": [
    "/api/v1/chat/completions",
    "/api/v1/completions",
    "/api/v1/messages",
    "/api/v1/responses"
  ],
  "benchmarks": {
    "design_arena": [
      {
        "arena": "models",
        "category": "3d",
        "elo": 974,
        "win_rate": 21.1,
        "rank": 118
      },
      {
        "arena": "models",
        "category": "asciiart",
        "elo": 986,
        "win_rate": 24.2,
        "rank": 74
      },
      {
        "arena": "models",
        "category": "codecategories",
        "elo": 1019,
        "win_rate": 21.2,
        "rank": 125
      },
      {
        "arena": "models",
        "category": "dataviz",
        "elo": 1015,
        "win_rate": 22.9,
        "rank": 116
      },
      {
        "arena": "models",
        "category": "gamedev",
        "elo": 979,
        "win_rate": 18.4,
        "rank": 126
      },
      {
        "arena": "models",
        "category": "svg",
        "elo": 1001,
        "win_rate": 24.8,
        "rank": 90
      },
      {
        "arena": "models",
        "category": "uicomponent",
        "elo": 993,
        "win_rate": 19.6,
        "rank": 120
      },
      {
        "arena": "models",
        "category": "website",
        "elo": 1036,
        "win_rate": 22.2,
        "rank": 127
      }
    ],
    "artificial_analysis": {
      "intelligence_index": null,
      "coding_index": 31.1,
      "agentic_index": 1.6
    }
  },
  "reasoning": {
    "mandatory": false,
    "default_enabled": true,
    "supported_efforts": [
      "high",
      "medium",
      "low",
      "none"
    ],
    "default_effort": "medium"
  }
}
openrouter.ai/inception/mercury-2