cerebras.json

 1{
 2    "name": "Cerebras",
 3    "id": "cerebras",
 4    "type": "openai-compat",
 5    "api_key": "$CEREBRAS_API_KEY",
 6    "api_endpoint": "https://api.cerebras.ai/v1",
 7    "default_large_model_id": "gpt-oss-120b",
 8    "default_small_model_id": "qwen-3-235b-a22b-instruct-2507",
 9    "default_headers": {
10        "X-Cerebras-3rd-Party-Integration": "crush"
11    },
12    "models": [
13        {
14            "id": "gpt-oss-120b",
15            "name": "OpenAI GPT OSS",
16            "cost_per_1m_in": 0.35,
17            "cost_per_1m_out": 0.75,
18            "context_window": 131072,
19            "default_max_tokens": 25000,
20            "can_reason": true,
21            "reasoning_levels": [
22                "low",
23                "medium",
24                "high"
25            ],
26            "default_reasoning_efforts": "medium",
27            "supports_attachments": false
28        },
29        {
30            "id": "qwen-3-235b-a22b-instruct-2507",
31            "name": "Qwen 3 235B Instruct",
32            "cost_per_1m_in": 0.6,
33            "cost_per_1m_out": 1.2,
34            "context_window": 131072,
35            "default_max_tokens": 25000,
36            "can_reason": false,
37            "supports_attachments": false
38        },
39        {
40            "id": "zai-glm-4.7",
41            "name": "Z.ai GLM 4.7",
42            "cost_per_1m_in": 2.25,
43            "cost_per_1m_out": 2.75,
44            "context_window": 131072,
45            "default_max_tokens": 25000,
46            "can_reason": false,
47            "supports_attachments": false,
48            "temperature": 1,
49            "top_p": 0.95
50        }
51    ]
52}