Omni API
textplannedtext-generation-preview

Text Generation Preview

Preview catalog entry used to prove the Portal can render text-only model contracts. It is not exposed for public task submission until the text worker and billing policy are finalized.

Open PlaygroundCreate API key
Runtime facts
Release
text-generation-preview-2026-05-planned
Latency
Text generation worker integration is planned after image/video stability.
Pricing
Token pricing planned
Status
planned
Task types
text_generationchat_completion
Inputs
prompt (textarea) *

Text prompt or chat instruction.

Used by text_generation, chat_completion

system_prompt (textarea)

Optional system instruction for chat-style models.

Used by text_generation, chat_completion

max_tokens (number)

Maximum generated tokens.

Used by text_generation, chat_completion

temperature (number)

Sampling temperature.

Used by text_generation, chat_completion

thinking_mode (select)

Optional model-specific thinking mode. Default stays disabled for user-facing text tasks.

disabledenabled

Used by text_generation, chat_completion

Outputs
result_text (text, text)
usage_json (json, json)
Playground

Text prompt or chat instruction.

Optional system instruction for chat-style models.

Maximum generated tokens.

Sampling temperature.

Optional model-specific thinking mode. Default stays disabled for user-facing text tasks.

prompt is required

1 credits estimated

Token pricing planned

Result

text output will appear here.

Task status
Submit a Playground task to see queue, runtime, credits, and task ID.
Recent tasks
No recent Playground tasks found for this identity.
API preview
curl https://omniapi.net/api/omni/tasks/create \
  -H "Authorization: Bearer $OMNI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model_group": "text-generation-preview",
  "task_type": "text_generation",
  "input": {
    "prompt": "Summarize this product brief in three concise bullets.",
    "max_tokens": 1024,
    "temperature": 0.7,
    "thinking_mode": "disabled"
  }
}'