Speka
Start free
All models
embeddingNVIDIA

NV-Embed v1

High-accuracy general-purpose text embeddings for semantic search and clustering.

nvidia/nv-embed-v1
Running· verified 2 min ago · 269ms

What is NV-Embed v1?

NV-Embed v1 is NVIDIA's embedding model. High-accuracy general-purpose text embeddings for semantic search and clustering. It runs on Speka's OpenAI-compatible /v1/embeddings endpoint at $0.016 per 1M input tokens, using the model ID nvidia/nv-embed-v1.

Test it live

Live playgroundNV-Embed v1in-browser demo

Embed text with NV-Embed v1

Enter text to see a vector preview.

How to call NV-Embed v1 via the API

Speka exposes NV-Embed v1 on an OpenAI-compatible /v1/embeddings endpoint. Point any OpenAI SDK at https://speka.me/v1, set model to nvidia/nv-embed-v1, and your existing code runs unchanged.

quickstart.py
from openai import OpenAI

client = OpenAI(
    base_url="https://speka.me/v1",
    api_key="sk-speka-live-...",  # your Speka key
)

resp = client.embeddings.create(
    model="nvidia/nv-embed-v1",
    input=["The quick brown fox"],
)
print(resp.data[0].embedding[:8])
FAQ

Frequently asked questions

NV-Embed v1 is NVIDIA's embedding model. High-accuracy general-purpose text embeddings for semantic search and clustering. It is available on Speka's OpenAI-compatible API using the model ID nvidia/nv-embed-v1.
On Speka, NV-Embed v1 is priced at $0.016 per 1M input tokens. Pricing is per-token with no subscription or seat fee — usage is billed against your plan's monthly allowance, then at the same rate with no overage penalty. Rates are current as of June 2026; check the pricing page for updates.
Speka exposes NV-Embed v1 on an OpenAI-compatible /v1/embeddings endpoint. Set base_url to https://speka.me/v1, supply your Speka key, and set model to "nvidia/nv-embed-v1" in any OpenAI SDK — no other code changes are required.
NV-Embed v1 supports a 32,768-token context window — roughly 25,000 words of input, suitable for long documents, multi-turn agents, and large-context tasks.
Yes. NV-Embed v1 is served through Speka's OpenAI-compatible endpoint, so any application built on the OpenAI Python or Node.js SDK, LangChain, or LlamaIndex can call it by changing only the base URL and model ID.

More embedding models

NVIDIA logo

NV-EmbedQA E5 v5

NVIDIA
embedding

Robust retrieval embeddings tuned for question answering and RAG pipelines.

nvidia/nv-embedqa-e5-v5
Input / 1M
$0.01
Open model
NVIDIA logo

Llama Nemotron Embed 1B v2

NVIDIA
embedding

Current-generation NVIDIA retrieval embedding model — 2048-dimension vectors tuned for multilingual RAG.

nvidia/llama-nemotron-embed-1b-v2
Input / 1M
$0.012
Open model
NVIDIA logo

Nemotron 3 Embed 1B

NVIDIA
embedding

Nemotron 3 embedding model for semantic search, clustering and reranking-style similarity scoring.

nvidia/nemotron-3-embed-1b
Input / 1M
$0.012
Open model