MistralResponse

Defined in: adapters/mistral.ts:110

Mistral AI API response structure.

OpenAI-compatible response format with Mistral-specific finish reasons. Includes generated message and token usage statistics.

See

• MistralRequest
• MistralMessage

Example

const response: MistralResponse = {
  id: 'cmpl-123',
  object: 'chat.completion',
  created: 1677652288,
  model: 'mistral-small',
  choices: [{
    index: 0,
    message: {
      role: 'assistant',
      content: 'Hello! How can I help?'
    },
    finish_reason: 'stop'
  }],
  usage: {
    prompt_tokens: 10,
    completion_tokens: 6,
    total_tokens: 16
  }
};

Properties

choices

choices: object[]

Defined in: adapters/mistral.ts:124

Array of completion choices (usually one)

finish_reason

finish_reason: "stop" | "length" | "model_length" | null

Stop reason: ‘stop’, ‘length’, ‘model_length’, or null

index

index: number

Choice index

message

message: MistralMessage

Generated message

---

created

created: number

Defined in: adapters/mistral.ts:118

Unix timestamp of creation

---

id

id: string

Defined in: adapters/mistral.ts:112

Unique completion identifier

---

model

model: string

Defined in: adapters/mistral.ts:121

Model used for generation

---

object

object: "chat.completion"

Defined in: adapters/mistral.ts:115

Object type - always ‘chat.completion’

---

usage

usage: object

Defined in: adapters/mistral.ts:136

Token usage statistics

completion_tokens

completion_tokens: number

Tokens in completion

prompt_tokens

prompt_tokens: number

Tokens in prompt

total_tokens

total_tokens: number

Total tokens used