embedding
Nemotron 3 Embed 8B
Nemotron 3 Embed 8BNemotron 3 Embed 8B is a high-accuracy embedding model for semantic search, code retrieval, and AI agents across complex, large-scale datasets.
Model details
View repositoryExample usage
Input
1import os
2from openai import OpenAI
3
4# Replace the empty string with your model id below
5model_id = ""
6baseten_api_key = os.environ["BASETEN_API_KEY"]
7
8client = OpenAI(
9 base_url=f"https://model-{model_id}.api.baseten.co/environments/production/sync/v1",
10 api_key=os.getenv("BASETEN_API_KEY")
11)
12
13inputs = ["query: What is the capital of France?",
14 "passage: Paris is the capital of France."]
15
16# Call model endpoint
17resp = client.embeddings.create(
18 model="nvidia/Nemotron-3-Embed-8B-BF16",
19 input=inputs
20)
21
22# Print the output of the model
23print(resp.to_json())JSON output
1{
2 "data": [
3 {
4 "embedding": [
5 0.001,
6 0.001
7 ],
8 "index": 1,
9 "object": "embedding"
10 }
11 ],
12 "model": "nvidia/Nemotron-3-Embed-1B-BF16",
13 "object": "list",
14 "usage": {
15 "prompt_tokens": 19,
16 "total_tokens": 19,
17 "completion_tokens": 0,
18 "prompt_tokens_details": null
19 },
20 "id": "embd-xxxx",
21 "created": 1784142524
22}
