1{
  2    "name": "Cerebras",
  3    "id": "cerebras",
  4    "type": "openai",
  5    "api_key": "$CEREBRAS_API_KEY",
  6    "api_endpoint": "https://api.cerebras.ai/v1",
  7    "default_large_model_id": "qwen-3-coder-480b",
  8    "default_small_model_id": "qwen-3-32b",
  9    "models": [
 10        {
 11            "id": "llama-4-scout-17b-16e-instruct",
 12            "name": "Llama 4 Scout",
 13            "cost_per_1m_in": 0.65,
 14            "cost_per_1m_out": 0.85,
 15            "context_window": 32768,
 16            "default_max_tokens": 4000,
 17            "can_reason": false,
 18            "supports_attachments": false
 19        },
 20        {
 21            "id": "llama3.1-8b",
 22            "name": "Llama 3.1 8B",
 23            "cost_per_1m_in": 0.1,
 24            "cost_per_1m_out": 0.1,
 25            "context_window": 32768,
 26            "default_max_tokens": 4000,
 27            "can_reason": false,
 28            "supports_attachments": false
 29        },
 30        {
 31            "id": "llama-3.3-70b",
 32            "name": "Llama 3.3 70B",
 33            "cost_per_1m_in": 0.85,
 34            "cost_per_1m_out": 1.2,
 35            "context_window": 128000,
 36            "default_max_tokens": 4000,
 37            "can_reason": false,
 38            "supports_attachments": false
 39        },
 40        {
 41            "id": "gpt-oss-120b",
 42            "name": "gpt-oss-120b",
 43            "cost_per_1m_in": 0.4,
 44            "cost_per_1m_out": 0.8,
 45            "context_window": 128000,
 46            "default_max_tokens": 65536,
 47            "can_reason": true,
 48            "has_reasoning_efforts": true,
 49            "default_reasoning_efforts": "medium",
 50            "supports_attachments": false
 51        },
 52        {
 53            "id": "qwen-3-32b",
 54            "name": "Qwen 3 32B",
 55            "cost_per_1m_in": 0.4,
 56            "cost_per_1m_out": 0.8,
 57            "context_window": 128000,
 58            "default_max_tokens": 32768,
 59            "can_reason": false,
 60            "supports_attachments": false
 61        },
 62        {
 63            "id": "llama-4-maverick-17b-128e-instruct",
 64            "name": "Llama 4 Maverick",
 65            "cost_per_1m_in": 0.2,
 66            "cost_per_1m_out": 0.6,
 67            "context_window": 32768,
 68            "default_max_tokens": 4000,
 69            "can_reason": false,
 70            "supports_attachments": false
 71        },
 72        {
 73            "id": "qwen-3-235b-a22b-instruct-2507",
 74            "name": "Qwen 3 235B Instruct",
 75            "cost_per_1m_in": 0.6,
 76            "cost_per_1m_out": 1.2,
 77            "context_window": 131072,
 78            "default_max_tokens": 16384,
 79            "can_reason": false,
 80            "supports_attachments": false
 81        },
 82        {
 83            "id": "qwen-3-235b-a22b-thinking-2507",
 84            "name": "Qwen 3 235B Thinking",
 85            "cost_per_1m_in": 0.6,
 86            "cost_per_1m_out": 1.2,
 87            "context_window": 128000,
 88            "default_max_tokens": 32768,
 89            "can_reason": false,
 90            "supports_attachments": false
 91        },
 92        {
 93            "id": "qwen-3-coder-480b",
 94            "name": "Qwen 3 480B Coder",
 95            "cost_per_1m_in": 2.0,
 96            "cost_per_1m_out": 2.0,
 97            "context_window": 131072,
 98            "default_max_tokens": 65536,
 99            "can_reason": false,
100            "supports_attachments": false
101        }
102    ]
103}