Summary
The model gpt-5.6-cyber is listed on OpenAI's models page, pricing page, and has a dedicated detail page, but is not present in the Braintrust proxy model catalog.
GPT-5.6 Cyber is a specialized model in the GPT-5.6 family designed for cybersecurity applications. It supports text and image input, text output, reasoning tokens, and has a 400K context window (272K input / 128K output). It is available via the Responses API (v1/responses).
Verification checklist
Proposed catalog entry
{
"gpt-5.6-cyber": {
"format": "openai",
"flavor": "chat",
"multimodal": true,
"input_cost_per_mil_tokens": 12.5,
"output_cost_per_mil_tokens": 75,
"input_cache_read_cost_per_mil_tokens": 1.25,
"input_cache_write_cost_per_mil_tokens": 15.625,
"displayName": "GPT-5.6 Cyber",
"reasoning": true,
"max_input_tokens": 272000,
"max_output_tokens": 128000,
"available_providers": ["openai"]
}
}
Sources
Summary
The model
gpt-5.6-cyberis listed on OpenAI's models page, pricing page, and has a dedicated detail page, but is not present in the Braintrust proxy model catalog.GPT-5.6 Cyber is a specialized model in the GPT-5.6 family designed for cybersecurity applications. It supports text and image input, text output, reasoning tokens, and has a 400K context window (272K input / 128K output). It is available via the Responses API (
v1/responses).Verification checklist
model_list.jsonforgpt-5.6-cyber— no match found; searched closed issues — no prior issue existsgpt-5.6-cyberfollows the established OpenAI naming convention (matches siblinggpt-5.6-solalready in catalog)Proposed catalog entry
{ "gpt-5.6-cyber": { "format": "openai", "flavor": "chat", "multimodal": true, "input_cost_per_mil_tokens": 12.5, "output_cost_per_mil_tokens": 75, "input_cache_read_cost_per_mil_tokens": 1.25, "input_cache_write_cost_per_mil_tokens": 15.625, "displayName": "GPT-5.6 Cyber", "reasoning": true, "max_input_tokens": 272000, "max_output_tokens": 128000, "available_providers": ["openai"] } }Sources