azure.json

  1{
  2  "name": "Azure OpenAI",
  3  "id": "azure",
  4  "type": "azure",
  5  "api_key": "$AZURE_OPENAI_API_KEY",
  6  "api_endpoint": "$AZURE_OPENAI_API_ENDPOINT",
  7  "default_large_model_id": "o4-mini",
  8  "default_small_model_id": "gpt-4o",
  9  "models": [
 10    {
 11      "id": "codex-mini-latest",
 12      "model": "Codex Mini",
 13      "cost_per_1m_in": 1.5,
 14      "cost_per_1m_out": 6,
 15      "cost_per_1m_in_cached": 0,
 16      "cost_per_1m_out_cached": 0.375,
 17      "context_window": 200000,
 18      "default_max_tokens": 50000,
 19      "can_reason": true,
 20      "has_reasoning_effort": true,
 21      "default_reasoning_effort": "medium",
 22      "supports_attachments": true
 23    },
 24    {
 25      "id": "o4-mini",
 26      "model": "o4 Mini",
 27      "cost_per_1m_in": 1.1,
 28      "cost_per_1m_out": 4.4,
 29      "cost_per_1m_in_cached": 0,
 30      "cost_per_1m_out_cached": 0.275,
 31      "context_window": 200000,
 32      "default_max_tokens": 50000,
 33      "can_reason": true,
 34      "has_reasoning_effort": true,
 35      "default_reasoning_effort": "medium",
 36      "supports_attachments": true
 37    },
 38    {
 39      "id": "o3",
 40      "model": "o3",
 41      "cost_per_1m_in": 2,
 42      "cost_per_1m_out": 8,
 43      "cost_per_1m_in_cached": 0,
 44      "cost_per_1m_out_cached": 0.5,
 45      "context_window": 200000,
 46      "default_max_tokens": 50000,
 47      "can_reason": true,
 48      "has_reasoning_effort": true,
 49      "default_reasoning_effort": "medium",
 50      "supports_attachments": true
 51    },
 52    {
 53      "id": "o3-pro",
 54      "model": "o3 Pro",
 55      "cost_per_1m_in": 20,
 56      "cost_per_1m_out": 80,
 57      "cost_per_1m_in_cached": 0,
 58      "cost_per_1m_out_cached": 0,
 59      "context_window": 200000,
 60      "default_max_tokens": 50000,
 61      "can_reason": true,
 62      "has_reasoning_effort": true,
 63      "default_reasoning_effort": "medium",
 64      "supports_attachments": true
 65    },
 66    {
 67      "id": "gpt-4.1",
 68      "model": "GPT-4.1",
 69      "cost_per_1m_in": 2,
 70      "cost_per_1m_out": 8,
 71      "cost_per_1m_in_cached": 0,
 72      "cost_per_1m_out_cached": 0.5,
 73      "context_window": 1047576,
 74      "default_max_tokens": 50000,
 75      "can_reason": false,
 76      "supports_attachments": true
 77    },
 78    {
 79      "id": "gpt-4.1-mini",
 80      "model": "GPT-4.1 Mini",
 81      "cost_per_1m_in": 0.39999999999999997,
 82      "cost_per_1m_out": 1.5999999999999999,
 83      "cost_per_1m_in_cached": 0,
 84      "cost_per_1m_out_cached": 0.09999999999999999,
 85      "context_window": 1047576,
 86      "default_max_tokens": 50000,
 87      "can_reason": false,
 88      "supports_attachments": true
 89    },
 90    {
 91      "id": "gpt-4.1-nano",
 92      "model": "GPT-4.1 Nano",
 93      "cost_per_1m_in": 0.09999999999999999,
 94      "cost_per_1m_out": 0.39999999999999997,
 95      "cost_per_1m_in_cached": 0,
 96      "cost_per_1m_out_cached": 0.024999999999999998,
 97      "context_window": 1047576,
 98      "default_max_tokens": 50000,
 99      "can_reason": false,
100      "supports_attachments": true
101    },
102    {
103      "id": "gpt-4.5-preview",
104      "model": "GPT-4.5 (Preview)",
105      "cost_per_1m_in": 75,
106      "cost_per_1m_out": 150,
107      "cost_per_1m_in_cached": 0,
108      "cost_per_1m_out_cached": 37.5,
109      "context_window": 128000,
110      "default_max_tokens": 50000,
111      "can_reason": false,
112      "supports_attachments": true
113    },
114    {
115      "id": "o3-mini",
116      "model": "o3 Mini",
117      "cost_per_1m_in": 1.1,
118      "cost_per_1m_out": 4.4,
119      "cost_per_1m_in_cached": 0,
120      "cost_per_1m_out_cached": 0.55,
121      "context_window": 200000,
122      "default_max_tokens": 50000,
123      "can_reason": true,
124      "has_reasoning_effort": true,
125      "default_reasoning_effort": "medium",
126      "supports_attachments": false
127    },
128    {
129      "id": "gpt-4o",
130      "model": "GPT-4o",
131      "cost_per_1m_in": 2.5,
132      "cost_per_1m_out": 10,
133      "cost_per_1m_in_cached": 0,
134      "cost_per_1m_out_cached": 1.25,
135      "context_window": 128000,
136      "default_max_tokens": 20000,
137      "can_reason": false,
138      "supports_attachments": true
139    },
140    {
141      "id": "gpt-4o-mini",
142      "model": "GPT-4o-mini",
143      "cost_per_1m_in": 0.15,
144      "cost_per_1m_out": 0.6,
145      "cost_per_1m_in_cached": 0,
146      "cost_per_1m_out_cached": 0.075,
147      "context_window": 128000,
148      "default_max_tokens": 20000,
149      "can_reason": false,
150      "reasoning_effort": "",
151      "supports_attachments": true
152    }
153  ]
154}