POST /api/v1/chat_generate
Request parameters, generation config, and response schema for chat and text generation.
Use this endpoint for chat, summarization, extraction, RAG, structured output, and function calling. The SDKs wrap it as addis.chat.completions.create; see the Text Generation guides.
Request parameters
These parameters go in the root of your JSON body.
Prop
Type
Generation config
Use generation_config to tune the creativity and length of the response.
Prop
Type
Response
The API returns a JSON object.
{
"response_text": "The generated text response...",
"finish_reason": "stop",
"usage_metadata": {
"prompt_token_count": 12,
"candidates_token_count": 45,
"total_token_count": 57
},
"modelVersion": "Addis-፩-አሌፍ"
}Prop
Type
Token Counting
Token counts vary by language, script, punctuation, and input structure. For Amharic, one word typically uses around 1.5 to 1.8 tokens. For the exact count, use usage_metadata returned in the REST API response or the corresponding usage value returned by the official SDK.
Edit on GitHub
Last updated on