Openai compatible apis provided by liquid.ai
- Input
curl --location 'https://labs.liquid.ai/api/v1/chat/completions' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer <api-key>' \
--data '{
"messages": [
{
"role": "user",
"content": "What is the largest animal on earth? Explain the aniaml features in detail. "
}
],
"model": "liquid-preview-0.1",
"max_tokens": 2048,
"top_p": 0.9,
"temperature": 0.9,
}'
- Output
{
"id": "cmpl-289eaf41592444718a4a3096d705e5d8",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "The largest animal on Earth is the Blue Whale (Balaenoptera musculus). It is a marine mammal belonging to the baleen whale group, characterized by its massive size, long and slender body, and absence of teeth.",
"role": "assistant",
"function_call": null,
"tool_calls": null
}
}
],
"created": 41593,
"model": "liquid-preview-0.1",
"object": "chat.completion",
"system_fingerprint": null,
"usage": {
"completion_tokens": 657,
"prompt_tokens": 29,
"total_tokens": 686
}
}
- Input You can adjust dimensions for the embedding output to trade accuracy with speed. The max dimension is 768.
curl --location 'https://labs.liquid.ai/api/v1/embeddings' \
--header 'Content-Type: application/json' \
--header 'X-API-Key: <api-key>' \
--data '{
"input": [
"My name is ",
"The president of the United States is ",
"The future of AI is "
],
"model": "liquid-embedding-0",
"encoding_format": "float",
"dimensions": 64
}'
- Output
{
"object": "list",
"data": [
{
"object": "embedding",
"embedding": [
0.0429396778345108,
-0.1442646086215973,
... total 64 elements
-0.014958425424993038,
0.21451351046562195
],
"index": 0
},
{
"object": "embedding",
"embedding": [
-0.014381556771695614,
0.15145191550254822,
... total 64 elements
-0.008231429383158684,
0.08554109185934067
],
"index": 1
},
{
"object": "embedding",
"embedding": [
-0.09036727249622345,
0.18065275251865387,
... total 64 elements
-0.014518562704324722,
0.10919401794672012
],
"index": 2
}
],
"model": "liquid-embedding-0",
"usage": {
"prompt_tokens": 21,
"total_tokens": 21
}
}