Models

Every request is served by a Nebula model. BreachLine routes each request across its engine matrix and does not expose the underlying engines, so model ids are BreachLine product names.

Model catalog

Model	Use it for	Endpoints	Context
nebula-4.5	Default text model: reasoning, tool calling and security analysis. Smart-routed per task.	Chat completions, Responses, Messages, Batches	256,000 tokens
nebula-v	Images and visual documents. Listed only while a vision engine is enabled.	Chat completions	Reported by GET /v1/models
nebula-embed	768-dimensional text embeddings for retrieval.	Embeddings	8,192 tokens
nebula-rerank	Relevance ranking of documents against a query.	Rerank

Choose a model

• Start with nebula-4.5 for chat, agents and structured output.
• Send images with nebula-v using image_url content parts.
• Use nebula-embed with task_type set to RETRIEVAL_DOCUMENT when indexing and RETRIEVAL_QUERY when searching.
• Use nebula-rerank to order search results before you show them or pass them to a model.

Reasoning effort

nebula-4.5 accepts reasoning_effort on chat completions: none, low, medium, high, max, xhigh or ultra. Higher values request a more capable engine and deeper reasoning, which costs more tokens. Omit it to let routing adapt to each request.

Unknown and retired ids

Any id that is not a current Nebula model, including retired Nebula ids and vendor model names, resolves to nebula-4.5. Requests do not fail because of an old id, but you should update it: the response’s model field tells you what served the request.

List models at run time

GET /api/v1/llm/v1/models returns the models available to your key, each with its context length, capabilities and the GBP price per million tokens. The response is cached privately for 60 seconds.

List modelsCopy

curl https://api.breachline.io/api/v1/llm/v1/models \
  -H "Authorization: Bearer $BREACHLINE_API_KEY"

The pricing values below are placeholders; read the live values from the endpoint.

Response (abridged)Copy

{
  "object": "list",
  "data": [
    {
      "id": "nebula-4.5",
      "object": "model",
      "owned_by": "breachline",
      "capabilities": ["chat", "completions", "reasoning", "tool_calling", "function_calling", "security_analysis"],
      "context_length": 256000,
      "pricing": { "input": 0.0, "output": 0.0, "currency": "GBP", "unit": "per_1m_tokens" }
    }
  ]
}