llm.rb 12.0.0 → 12.2.0

data/data/deepinfra.json

@@ -10,10 +10,12 @@
     "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": {
       "id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
       "name": "Llama 4 Maverick 17B FP8",
+      "description": "Open multimodal Llama model for strong reasoning and fast responses",
       "family": "llama",
-      "attachment": false,
+      "attachment": true,
       "reasoning": false,
       "tool_call": false,
+      "structured_output": true,
       "release_date": "2025-04-05",
       "last_updated": "2025-04-05",
       "modalities": {
@@ -38,10 +40,12 @@
     "meta-llama/Llama-4-Scout-17B-16E-Instruct": {
       "id": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
       "name": "Llama 4 Scout 17B",
+      "description": "Open multimodal Llama model for long-context analysis and efficient agents",
       "family": "llama",
-      "attachment": false,
+      "attachment": true,
       "reasoning": false,
       "tool_call": true,
+      "structured_output": true,
       "release_date": "2025-04-05",
       "last_updated": "2025-04-05",
       "modalities": {
@@ -66,10 +70,12 @@
     "meta-llama/Llama-3.3-70B-Instruct-Turbo": {
       "id": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
       "name": "Llama 3.3 70B Turbo",
+      "description": "Compact Llama instruction model for fast chat and local deployment",
       "family": "llama",
       "attachment": false,
       "reasoning": false,
       "tool_call": true,
+      "structured_output": true,
       "release_date": "2024-12-06",
       "last_updated": "2024-12-06",
       "modalities": {
@@ -93,6 +99,7 @@
     "moonshotai/Kimi-K2.6": {
       "id": "moonshotai/Kimi-K2.6",
       "name": "Kimi K2.6",
+      "description": "Kimi multimodal agent model for visual understanding, coding, and planning",
       "family": "kimi-k2",
       "attachment": true,
       "reasoning": true,
@@ -134,6 +141,7 @@
     "moonshotai/Kimi-K2.5": {
       "id": "moonshotai/Kimi-K2.5",
       "name": "Kimi K2.5",
+      "description": "Kimi multimodal agent model for visual understanding, coding, and planning",
       "family": "kimi-k2",
       "attachment": true,
       "reasoning": true,
@@ -172,9 +180,52 @@
         "cache_read": 0.07
       }
     },
+    "moonshotai/Kimi-K2.7-Code": {
+      "id": "moonshotai/Kimi-K2.7-Code",
+      "name": "Kimi K2.7 Code",
+      "description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking",
+      "family": "kimi-k2",
+      "attachment": true,
+      "reasoning": true,
+      "reasoning_options": [
+        {
+          "type": "toggle"
+        }
+      ],
+      "tool_call": true,
+      "interleaved": {
+        "field": "reasoning_content"
+      },
+      "structured_output": true,
+      "temperature": false,
+      "knowledge": "2025-01",
+      "release_date": "2026-06-12",
+      "last_updated": "2026-06-12",
+      "modalities": {
+        "input": [
+          "text",
+          "image",
+          "video"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 262144,
+        "output": 262144
+      },
+      "cost": {
+        "input": 0.74,
+        "output": 3.5,
+        "cache_read": 0.15
+      }
+    },
     "google/gemma-4-31B-it": {
       "id": "google/gemma-4-31B-it",
       "name": "Gemma 4 31B IT",
+      "description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning",
       "family": "gemma",
       "attachment": true,
       "reasoning": true,
@@ -191,7 +242,8 @@
       "modalities": {
         "input": [
           "text",
-          "image"
+          "image",
+          "video"
         ],
         "output": [
           "text"
@@ -210,6 +262,7 @@
     "google/gemma-4-26B-A4B-it": {
       "id": "google/gemma-4-26B-A4B-it",
       "name": "Gemma 4 26B A4B IT",
+      "description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
       "family": "gemma",
       "attachment": true,
       "reasoning": true,
@@ -245,11 +298,13 @@
     "Qwen/Qwen3.6-35B-A3B": {
       "id": "Qwen/Qwen3.6-35B-A3B",
       "name": "Qwen3.6 35B A3B",
+      "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
       "family": "qwen",
       "attachment": true,
       "reasoning": true,
       "reasoning_options": [],
       "tool_call": true,
+      "structured_output": true,
       "temperature": true,
       "release_date": "2026-04-01",
       "last_updated": "2026-04-01",
@@ -273,14 +328,184 @@
         "output": 0.95
       }
     },
+    "Qwen/Qwen3.7-Max": {
+      "id": "Qwen/Qwen3.7-Max",
+      "name": "Qwen3.7 Max",
+      "description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks",
+      "family": "qwen",
+      "attachment": false,
+      "reasoning": false,
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": true,
+      "release_date": "2026-05-21",
+      "last_updated": "2026-05-21",
+      "modalities": {
+        "input": [
+          "text"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": false,
+      "limit": {
+        "context": 256000,
+        "output": 65536
+      },
+      "cost": {
+        "input": 2.5,
+        "output": 7.5,
+        "cache_read": 0.5,
+        "tiers": [
+          {
+            "input": 5,
+            "output": 15,
+            "cache_read": 1,
+            "tier": {
+              "type": "context",
+              "size": 32000
+            }
+          },
+          {
+            "input": 6.25,
+            "output": 18.5,
+            "cache_read": 1.25,
+            "tier": {
+              "type": "context",
+              "size": 128000
+            }
+          }
+        ]
+      }
+    },
+    "Qwen/Qwen3.6-27B": {
+      "id": "Qwen/Qwen3.6-27B",
+      "name": "Qwen3.6 27B",
+      "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
+      "family": "qwen",
+      "attachment": true,
+      "reasoning": true,
+      "reasoning_options": [],
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": true,
+      "release_date": "2026-04-22",
+      "last_updated": "2026-04-22",
+      "modalities": {
+        "input": [
+          "text",
+          "image",
+          "video",
+          "audio"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 262144,
+        "output": 65536
+      },
+      "cost": {
+        "input": 0.32,
+        "output": 3.2
+      }
+    },
+    "Qwen/Qwen3-Next-80B-A3B-Instruct": {
+      "id": "Qwen/Qwen3-Next-80B-A3B-Instruct",
+      "name": "Qwen3-Next 80B-A3B Instruct",
+      "description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
+      "family": "qwen",
+      "attachment": false,
+      "reasoning": false,
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": true,
+      "knowledge": "2025-04",
+      "release_date": "2025-09",
+      "last_updated": "2025-09",
+      "modalities": {
+        "input": [
+          "text"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 262144,
+        "output": 32768
+      },
+      "cost": {
+        "input": 0.09,
+        "output": 1.1
+      }
+    },
+    "Qwen/Qwen3-Max": {
+      "id": "Qwen/Qwen3-Max",
+      "name": "Qwen3 Max",
+      "description": "Flagship Qwen3 model for coding agents, complex reasoning, and tool use",
+      "family": "qwen",
+      "attachment": false,
+      "reasoning": false,
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": true,
+      "knowledge": "2025-04",
+      "release_date": "2025-09-23",
+      "last_updated": "2025-09-23",
+      "modalities": {
+        "input": [
+          "text"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": false,
+      "limit": {
+        "context": 256000,
+        "output": 65536
+      },
+      "cost": {
+        "input": 1.2,
+        "output": 6,
+        "cache_read": 0.24,
+        "tiers": [
+          {
+            "input": 2.4,
+            "output": 12,
+            "cache_read": 0.48,
+            "tier": {
+              "type": "context",
+              "size": 32000
+            }
+          },
+          {
+            "input": 3,
+            "output": 15,
+            "cache_read": 0.6,
+            "tier": {
+              "type": "context",
+              "size": 128000
+            }
+          }
+        ]
+      }
+    },
     "Qwen/Qwen3.5-397B-A17B": {
       "id": "Qwen/Qwen3.5-397B-A17B",
       "name": "Qwen 3.5 397B A17B",
+      "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
       "family": "qwen",
       "attachment": true,
       "reasoning": true,
       "reasoning_options": [],
       "tool_call": true,
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2025-01",
       "release_date": "2026-02-01",
@@ -306,13 +531,148 @@
         "cache_read": 0.22
       }
     },
+    "Qwen/Qwen3.5-122B-A10B": {
+      "id": "Qwen/Qwen3.5-122B-A10B",
+      "name": "Qwen3.5 122B-A10B",
+      "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
+      "family": "qwen",
+      "attachment": true,
+      "reasoning": true,
+      "reasoning_options": [],
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": true,
+      "release_date": "2026-02-23",
+      "last_updated": "2026-02-23",
+      "modalities": {
+        "input": [
+          "text",
+          "image",
+          "video",
+          "audio"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 262144,
+        "output": 65536
+      },
+      "cost": {
+        "input": 0.29,
+        "output": 2.4
+      }
+    },
+    "Qwen/Qwen3.5-27B": {
+      "id": "Qwen/Qwen3.5-27B",
+      "name": "Qwen3.5 27B",
+      "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
+      "family": "qwen",
+      "attachment": true,
+      "reasoning": true,
+      "reasoning_options": [],
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": true,
+      "release_date": "2026-02-23",
+      "last_updated": "2026-02-23",
+      "modalities": {
+        "input": [
+          "text",
+          "image",
+          "video",
+          "audio"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 262144,
+        "output": 65536
+      },
+      "cost": {
+        "input": 0.26,
+        "output": 2.6
+      }
+    },
+    "Qwen/Qwen3.5-9B": {
+      "id": "Qwen/Qwen3.5-9B",
+      "name": "Qwen3.5 9B",
+      "description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
+      "family": "qwen",
+      "attachment": true,
+      "reasoning": true,
+      "reasoning_options": [],
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": true,
+      "release_date": "2026-02-23",
+      "last_updated": "2026-02-23",
+      "modalities": {
+        "input": [
+          "text",
+          "image",
+          "video"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 262144,
+        "output": 65536
+      },
+      "cost": {
+        "input": 0.1,
+        "output": 0.15
+      }
+    },
+    "Qwen/Qwen3-32B": {
+      "id": "Qwen/Qwen3-32B",
+      "name": "Qwen3 32B",
+      "description": "Dense open Qwen model for self-hosted chat, reasoning, and coding",
+      "family": "qwen",
+      "attachment": false,
+      "reasoning": true,
+      "reasoning_options": [],
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": true,
+      "knowledge": "2025-04",
+      "release_date": "2025-04",
+      "last_updated": "2025-04",
+      "modalities": {
+        "input": [
+          "text"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 40960,
+        "output": 16384
+      },
+      "cost": {
+        "input": 0.08,
+        "output": 0.28
+      }
+    },
     "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo": {
       "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo",
       "name": "Qwen3 Coder 480B A35B Instruct Turbo",
+      "description": "Qwen coding model for software agents, repository edits, and code reasoning",
       "family": "qwen",
       "attachment": false,
       "reasoning": false,
       "tool_call": true,
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2025-04",
       "release_date": "2025-07-23",
@@ -332,17 +692,20 @@
       },
       "cost": {
         "input": 0.3,
-        "output": 1
+        "output": 1,
+        "cache_read": 0.1
       }
     },
     "Qwen/Qwen3.5-35B-A3B": {
       "id": "Qwen/Qwen3.5-35B-A3B",
       "name": "Qwen 3.5 35B A3B",
+      "description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
       "family": "qwen",
       "attachment": true,
       "reasoning": true,
       "reasoning_options": [],
       "tool_call": true,
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2025-01",
       "release_date": "2026-02-01",
@@ -371,6 +734,7 @@
     "openai/gpt-oss-120b": {
       "id": "openai/gpt-oss-120b",
       "name": "GPT OSS 120B",
+      "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads",
       "family": "gpt-oss",
       "attachment": false,
       "reasoning": true,
@@ -385,6 +749,7 @@
         }
       ],
       "tool_call": true,
+      "structured_output": true,
       "temperature": true,
       "release_date": "2025-08-05",
       "last_updated": "2025-08-05",
@@ -402,13 +767,14 @@
         "output": 16384
       },
       "cost": {
-        "input": 0.039,
-        "output": 0.19
+        "input": 0.037,
+        "output": 0.17
       }
     },
     "openai/gpt-oss-20b": {
       "id": "openai/gpt-oss-20b",
       "name": "GPT OSS 20B",
+      "description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads",
       "family": "gpt-oss",
       "attachment": false,
       "reasoning": true,
@@ -423,6 +789,7 @@
         }
       ],
       "tool_call": true,
+      "structured_output": true,
       "temperature": true,
       "release_date": "2025-08-05",
       "last_updated": "2025-08-05",
@@ -447,6 +814,7 @@
     "XiaomiMiMo/MiMo-V2.5": {
       "id": "XiaomiMiMo/MiMo-V2.5",
       "name": "MiMo-V2.5",
+      "description": "Open MiMo model for multimodal coding agents and long-context automation",
       "family": "mimo",
       "attachment": true,
       "reasoning": true,
@@ -459,6 +827,7 @@
       "interleaved": {
         "field": "reasoning_content"
       },
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2024-12",
       "release_date": "2026-04-22",
@@ -482,30 +851,15 @@
       "cost": {
         "input": 0.4,
         "output": 2,
-        "cache_read": 0.08,
-        "tiers": [
-          {
-            "input": 0.8,
-            "output": 4,
-            "cache_read": 0.16,
-            "tier": {
-              "type": "context",
-              "size": 256000
-            }
-          }
-        ],
-        "context_over_200k": {
-          "input": 0.8,
-          "output": 4,
-          "cache_read": 0.16
-        }
+        "cache_read": 0.08
       }
     },
     "XiaomiMiMo/MiMo-V2.5-Pro": {
       "id": "XiaomiMiMo/MiMo-V2.5-Pro",
       "name": "MiMo-V2.5-Pro",
+      "description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution",
       "family": "mimo",
-      "attachment": false,
+      "attachment": true,
       "reasoning": true,
       "reasoning_options": [
         {
@@ -516,13 +870,15 @@
       "interleaved": {
         "field": "reasoning_content"
       },
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2024-12",
       "release_date": "2026-04-22",
       "last_updated": "2026-04-22",
       "modalities": {
         "input": [
-          "text"
+          "text",
+          "audio"
         ],
         "output": [
           "text"
@@ -536,28 +892,112 @@
       "cost": {
         "input": 1,
         "output": 3,
-        "cache_read": 0.2,
-        "tiers": [
-          {
-            "input": 2,
-            "output": 6,
-            "cache_read": 0.4,
-            "tier": {
-              "type": "context",
-              "size": 256000
-            }
-          }
+        "cache_read": 0.2
+      }
+    },
+    "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning": {
+      "id": "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning",
+      "name": "Nemotron 3 Nano Omni 30B A3B Reasoning",
+      "description": "Open Nemotron omni model combining reasoning with text, vision, and audio",
+      "family": "nemotron",
+      "attachment": true,
+      "reasoning": true,
+      "reasoning_options": [],
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": true,
+      "release_date": "2026-04-28",
+      "last_updated": "2026-04-28",
+      "modalities": {
+        "input": [
+          "text",
+          "image",
+          "video",
+          "audio"
         ],
-        "context_over_200k": {
-          "input": 2,
-          "output": 6,
-          "cache_read": 0.4
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 262144,
+        "output": 65536
+      },
+      "cost": {
+        "input": 0.2,
+        "output": 0.8
+      }
+    },
+    "nvidia/Nemotron-3-Nano-30B-A3B": {
+      "id": "nvidia/Nemotron-3-Nano-30B-A3B",
+      "name": "Nemotron 3 Nano 30B A3B",
+      "description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents",
+      "family": "nemotron",
+      "attachment": false,
+      "reasoning": true,
+      "reasoning_options": [
+        {
+          "type": "toggle"
         }
+      ],
+      "tool_call": true,
+      "temperature": true,
+      "release_date": "2025-12-15",
+      "last_updated": "2025-12-15",
+      "modalities": {
+        "input": [
+          "text"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 262144,
+        "output": 262144
+      },
+      "cost": {
+        "input": 0.05,
+        "output": 0.2
+      }
+    },
+    "nvidia/Llama-3.3-Nemotron-Super-49B-v1.5": {
+      "id": "nvidia/Llama-3.3-Nemotron-Super-49B-v1.5",
+      "name": "Llama 3.3 Nemotron Super 49B v1.5",
+      "description": "Nemotron model for efficient reasoning, coding, and specialized AI agents",
+      "family": "nemotron",
+      "attachment": false,
+      "reasoning": true,
+      "reasoning_options": [],
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": true,
+      "release_date": "2025-07-25",
+      "last_updated": "2025-07-25",
+      "modalities": {
+        "input": [
+          "text"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 131072,
+        "output": 131072
+      },
+      "cost": {
+        "input": 0.4,
+        "output": 0.4
       }
     },
     "zai-org/GLM-4.7-Flash": {
       "id": "zai-org/GLM-4.7-Flash",
       "name": "GLM-4.7-Flash",
+      "description": "Efficient GLM model for fast reasoning, coding, and agent workflows",
       "family": "glm-flash",
       "attachment": false,
       "reasoning": true,
@@ -566,6 +1006,7 @@
       "interleaved": {
         "field": "reasoning_content"
       },
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2025-04",
       "release_date": "2026-01-19",
@@ -585,12 +1026,14 @@
       },
       "cost": {
         "input": 0.06,
-        "output": 0.4
+        "output": 0.4,
+        "cache_read": 0.01
       }
     },
     "zai-org/GLM-4.6": {
       "id": "zai-org/GLM-4.6",
       "name": "GLM-4.6",
+      "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
       "family": "glm",
       "attachment": false,
       "reasoning": true,
@@ -603,6 +1046,7 @@
       "interleaved": {
         "field": "reasoning_content"
       },
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2025-04",
       "release_date": "2025-09-30",
@@ -629,6 +1073,7 @@
     "zai-org/GLM-5": {
       "id": "zai-org/GLM-5",
       "name": "GLM-5",
+      "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
       "family": "glm",
       "attachment": false,
       "reasoning": true,
@@ -641,6 +1086,7 @@
       "interleaved": {
         "field": "reasoning_content"
       },
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2025-12",
       "release_date": "2026-02-12",
@@ -667,6 +1113,7 @@
     "zai-org/GLM-4.7": {
       "id": "zai-org/GLM-4.7",
       "name": "GLM-4.7",
+      "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
       "family": "glm",
       "attachment": false,
       "reasoning": true,
@@ -679,6 +1126,7 @@
       "interleaved": {
         "field": "reasoning_content"
       },
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2025-04",
       "release_date": "2025-12-22",
@@ -705,6 +1153,7 @@
     "zai-org/GLM-5.2": {
       "id": "zai-org/GLM-5.2",
       "name": "GLM-5.2",
+      "description": "Open flagship GLM for long-horizon coding agents and million-token context work",
       "family": "glm",
       "attachment": false,
       "reasoning": true,
@@ -744,7 +1193,7 @@
         "output": 32768
       },
       "cost": {
-        "input": 0.95,
+        "input": 0.93,
         "output": 3,
         "cache_read": 0.18
       }
@@ -752,6 +1201,7 @@
     "zai-org/GLM-5.1": {
       "id": "zai-org/GLM-5.1",
       "name": "GLM-5.1",
+      "description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
       "family": "glm",
       "attachment": false,
       "reasoning": true,
@@ -791,6 +1241,7 @@
     "deepseek-ai/DeepSeek-R1-0528": {
       "id": "deepseek-ai/DeepSeek-R1-0528",
       "name": "DeepSeek-R1-0528",
+      "description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools",
       "attachment": false,
       "reasoning": true,
       "reasoning_options": [],
@@ -798,6 +1249,7 @@
       "interleaved": {
         "field": "reasoning_content"
       },
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2024-07",
       "release_date": "2025-05-28",
@@ -824,6 +1276,7 @@
     "deepseek-ai/DeepSeek-V4-Flash": {
       "id": "deepseek-ai/DeepSeek-V4-Flash",
       "name": "DeepSeek V4 Flash",
+      "description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work",
       "family": "deepseek-flash",
       "attachment": false,
       "reasoning": true,
@@ -864,14 +1317,15 @@
         "output": 16384
       },
       "cost": {
-        "input": 0.1,
-        "output": 0.2,
-        "cache_read": 0.02
+        "input": 0.09,
+        "output": 0.18,
+        "cache_read": 0.018
       }
     },
     "deepseek-ai/DeepSeek-V4-Pro": {
       "id": "deepseek-ai/DeepSeek-V4-Pro",
       "name": "DeepSeek V4 Pro",
+      "description": "Open MoE flagship with million-token context for coding and long agent runs",
       "family": "deepseek-thinking",
       "attachment": false,
       "reasoning": true,
@@ -920,6 +1374,7 @@
     "deepseek-ai/DeepSeek-V3.2": {
       "id": "deepseek-ai/DeepSeek-V3.2",
       "name": "DeepSeek-V3.2",
+      "description": "DeepSeek chat model for instruction following, coding, and analysis",
       "attachment": false,
       "reasoning": true,
       "reasoning_options": [
@@ -931,6 +1386,7 @@
       "interleaved": {
         "field": "reasoning_content"
       },
+      "structured_output": true,
       "temperature": true,
       "knowledge": "2024-12",
       "release_date": "2025-12-02",
@@ -957,6 +1413,7 @@
     "MiniMaxAI/MiniMax-M2.5": {
       "id": "MiniMaxAI/MiniMax-M2.5",
       "name": "MiniMax M2.5",
+      "description": "MiniMax model for chat, coding, office work, and agentic tasks",
       "family": "minimax",
       "attachment": false,
       "reasoning": true,
@@ -985,8 +1442,71 @@
       "cost": {
         "input": 0.15,
         "output": 1.15,
-        "cache_read": 0.03,
-        "cache_write": 0.375
+        "cache_read": 0.03
+      }
+    },
+    "MiniMaxAI/MiniMax-M3": {
+      "id": "MiniMaxAI/MiniMax-M3",
+      "name": "MiniMax-M3",
+      "description": "MiniMax multimodal model for long-context coding, perception, and agent planning",
+      "family": "minimax",
+      "attachment": true,
+      "reasoning": true,
+      "reasoning_options": [],
+      "tool_call": true,
+      "temperature": true,
+      "release_date": "2026-06-01",
+      "last_updated": "2026-06-01",
+      "modalities": {
+        "input": [
+          "text",
+          "image",
+          "video"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 524288,
+        "output": 128000
+      },
+      "cost": {
+        "input": 0.3,
+        "output": 1.2,
+        "cache_read": 0.06
+      }
+    },
+    "MiniMaxAI/MiniMax-M2.7": {
+      "id": "MiniMaxAI/MiniMax-M2.7",
+      "name": "MiniMax-M2.7",
+      "description": "Open MiniMax flagship for coding agents, office automation, and complex environments",
+      "family": "minimax",
+      "attachment": false,
+      "reasoning": true,
+      "reasoning_options": [],
+      "tool_call": true,
+      "temperature": true,
+      "release_date": "2026-03-18",
+      "last_updated": "2026-03-18",
+      "modalities": {
+        "input": [
+          "text"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": true,
+      "limit": {
+        "context": 196608,
+        "output": 131072
+      },
+      "cost": {
+        "input": 0.25,
+        "output": 1,
+        "cache_read": 0.05
       }
     }
   }