OllamaResponse

Defined in: adapters/ollama.ts:109

Ollama API response structure.

Includes the generated message plus performance metrics like duration and token counts specific to local model execution.

See

• OllamaRequest
• OllamaMessage

Example

const response: OllamaResponse = {
  model: 'llama2',
  created_at: '2024-01-15T12:00:00Z',
  message: {
    role: 'assistant',
    content: 'Hello! How can I help?'
  },
  done: true,
  total_duration: 1500000000,
  prompt_eval_count: 10,
  eval_count: 6
};

Properties

created_at

created_at: string

Defined in: adapters/ollama.ts:114

ISO 8601 timestamp

---

done

done: boolean

Defined in: adapters/ollama.ts:120

Whether generation is complete

---

eval_count?

optional eval_count: number

Defined in: adapters/ollama.ts:132

Number of tokens in generation

---

load_duration?

optional load_duration: number

Defined in: adapters/ollama.ts:126

Model loading duration in nanoseconds

---

message

message: OllamaMessage

Defined in: adapters/ollama.ts:117

Generated message

---

model

model: string

Defined in: adapters/ollama.ts:111

Model that generated the response

---

prompt_eval_count?

optional prompt_eval_count: number

Defined in: adapters/ollama.ts:129

Number of tokens in prompt evaluation

---

total_duration?

optional total_duration: number

Defined in: adapters/ollama.ts:123

Total inference duration in nanoseconds