Documentation Index
Fetch the complete documentation index at: https://docs.moodmnky.com/llms.txt
Use this file to discover all available pages before exploring further.
Text Generation
This guide covers text generation capabilities of the Ollama service.
Generation Endpoint
Generate text using AI models.
Generation Endpoints
Generate Text
POST /api/generate
curl -X POST "https://ollama.moodmnky.com/api/generate" \
-H "x-api-key: your_api_key" \
-H "Content-Type: application/json" \
-d '{
"model": "llama2",
"prompt": "Write a story about a robot learning to paint.",
"parameters": {
"temperature": 0.7,
"max_tokens": 500
}
}'
Stream Generation
POST /api/generate/stream
curl -X POST "https://ollama.moodmnky.com/api/generate/stream" \
-H "x-api-key: your_api_key" \
-H "Content-Type: application/json" \
-d '{
"model": "llama2",
"prompt": "Write a story about a robot learning to paint.",
"parameters": {
"temperature": 0.7,
"max_tokens": 500
}
}'
Generation Parameters
Core Parameters
| Parameter | Type | Description | Default |
|---|
| model | string | Model to use | Required |
| prompt | string | Input text | Required |
| temperature | float | Randomness (0.0-1.0) | 0.8 |
| max_tokens | integer | Maximum tokens to generate | 2048 |
| top_p | float | Nucleus sampling threshold | 0.9 |
| top_k | integer | Top-k sampling threshold | 40 |
| repeat_penalty | float | Repetition penalty | 1.1 |
| presence_penalty | float | Presence penalty | 0.0 |
| frequency_penalty | float | Frequency penalty | 0.0 |
Advanced Parameters
| Parameter | Type | Description | Default |
|---|
| stop_sequences | string[] | Sequences to stop generation | [] |
| seed | integer | Random seed for reproducibility | null |
| num_ctx | integer | Context window size | 2048 |
| num_predict | integer | Number of tokens to predict | -1 |
Standard Response
{
"text": "Generated text response",
"usage": {
"prompt_tokens": 10,
"completion_tokens": 100,
"total_tokens": 110
},
"model": "llama2",
"created_at": "2024-04-05T12:00:00Z"
}
Stream Response
{
"text": "Partial",
"usage": {
"completion_tokens": 1
}
}
// ... more chunks ...
{
"text": " response.",
"usage": {
"completion_tokens": 1
},
"done": true
}
Generation Examples
Basic Text Generation
const response = await client.ollama.generate({
model: "llama2",
prompt: "Write a story about a robot learning to paint.",
parameters: {
temperature: 0.7,
max_tokens: 500
}
});
console.log(response.text);
Streaming Generation
const stream = await client.ollama.generateStream({
model: "llama2",
prompt: "Write a story about a robot learning to paint.",
parameters: {
temperature: 0.7,
max_tokens: 500
}
});
for await (const chunk of stream) {
process.stdout.write(chunk.text);
}
Chat Completion
const response = await client.ollama.generate({
model: "llama2",
prompt: [
{ role: "system", content: "You are a helpful assistant." },
{ role: "user", content: "Tell me about painting techniques." }
],
parameters: {
temperature: 0.7
}
});
Best Practices
-
Prompt Engineering
- Be specific and clear in prompts
- Provide context when needed
- Use system messages for behavior control
- Structure multi-turn conversations properly
-
Parameter Tuning
- Lower temperature for factual responses
- Higher temperature for creative tasks
- Adjust max_tokens based on expected length
- Use stop sequences to control output
-
Performance Optimization
- Use streaming for long responses
- Batch similar requests when possible
- Cache frequently used responses
- Monitor token usage
Error Handling
try {
const response = await client.ollama.generate({
model: "llama2",
prompt: "Hello, world!"
});
} catch (error) {
switch (error.code) {
case "CONTEXT_LENGTH_EXCEEDED":
// Handle prompt too long
break;
case "RATE_LIMIT_EXCEEDED":
// Handle rate limiting
break;
case "INVALID_PARAMETERS":
// Handle invalid parameters
break;
}
}
Support & Resources
