Skip to main content

​
Overview

List all embedding-capable models available through RedPill. This endpoint returns models specifically designed for generating vector embeddings from text, images, or videos.
Embedding models are filtered separately from chat/completion models. Use GET /v1/models for chat models or GET /v1/embeddings/models for embedding models.

​
List All Embedding Models

curl https://api.redpill.ai/v1/embeddings/models \
  -H "Authorization: Bearer $REDPILL_API_KEY"

{
  "object": "list",
  "data": [
    {
      "id": "openai/text-embedding-3-large",
      "name": "OpenAI: Text Embedding 3 Large",
      "created": 1704067200,
      "input_modalities": ["text", "embeddings"],
      "output_modalities": ["embeddings"],
      "context_length": 8192,
      "max_output_length": 8192,
      "pricing": {
        "prompt": "0.00000013",
        "completion": "0",
        "image": "0",
        "request": "0"
      },
      "supported_sampling_parameters": ["temperature"],
      "supported_features": [],
      "description": "text-embedding-3-large is OpenAI's most capable embedding model for both english and non-english tasks. Embeddings are a numerical representation of text that can be used to measure the relatedness between two pieces of text."
    },
    {
      "id": "openai/text-embedding-3-small",
      "name": "OpenAI: Text Embedding 3 Small",
      "created": 1704067200,
      "input_modalities": ["text", "embeddings"],
      "output_modalities": ["embeddings"],
      "context_length": 8192,
      "max_output_length": 8192,
      "pricing": {
        "prompt": "0.00000002",
        "completion": "0"
      },
      "description": "Improved embedding model with better performance than ada-002"
    },
    {
      "id": "openai/text-embedding-ada-002",
      "name": "OpenAI: Text Embedding Ada 002",
      "created": 1704067200,
      "input_modalities": ["text", "embeddings"],
      "output_modalities": ["embeddings"],
      "context_length": 8192,
      "max_output_length": 8192,
      "pricing": {
        "prompt": "0.0000001",
        "completion": "0"
      },
      "description": "Legacy text embedding model"
    }
  ]
}

​
Response Fields

​
object
string
Always "list" for this endpoint
​
data
array
Array of embedding model objects

​
Filter by Provider

Get embedding models from a specific provider. 
curl https://api.redpill.ai/v1/embeddings/models/openai \
  -H "Authorization: Bearer $REDPILL_API_KEY"

​
URL Parameters

​
provider
string
required
Provider name to filter by (e.g., openai, voyage, jina, nomic, cohere)

​
Available Embedding Providers

ProviderModelsSpecializationMax Dimensions
OpenAItext-embedding-3-large, 3-small, ada-002General purpose3072
Voyagevoyage-3, voyage-3-liteSemantic search1024
Jinajina-embeddings-v3, jina-clip-v2Multi-modal1024
Nomicnomic-embed-text-v1.5, nomic-embed-videoText & video768
Cohereembed-english-v3.0, embed-multilingualMultilingual1024
Milvusmilvus-embeddingVector database768
Qdrantqdrant-embeddingVector search1536

Create Embeddings

Learn how to generate embeddings →

​
Model Selection Guide

​
By Use Case

Recommended: openai/text-embedding-3-large
  • Best overall quality
  • Supports 100+ languages
  • Customizable dimensions (256-3072)
Budget Option: openai/text-embedding-3-small
  • 5x cheaper
  • Good performance
  • Faster generation
Recommended: voyage/voyage-3
  • Optimized for retrieval tasks
  • Strong semantic understanding
  • Excellent for document search
Alternative: cohere/embed-english-v3.0
  • Built-in search/clustering modes
  • High accuracy for RAG applications
Recommended: jina/jina-clip-v2
  • Unified text and image embeddings
  • Same vector space for cross-modal search
  • Supports image URLs and base64
Alternative: openai/text-embedding-3-large
  • Supports text embeddings only
  • Use with vision models for image understanding
Recommended: nomic/nomic-embed-video-v1.5
  • Specialized for video content
  • Temporal understanding
  • Scene-level embeddings
Alternative: Extract frames and use jina/jina-clip-v2
Recommended: cohere/embed-multilingual-v3.0
  • 100+ languages
  • Strong cross-lingual performance
  • No language detection needed
Alternative: openai/text-embedding-3-large
  • Supports many languages
  • Consistent quality across languages
Recommended: openai/text-embedding-3-small
  • $0.02 per 1M tokens
  • Good quality/price ratio
Ultra-Budget: openai/text-embedding-ada-002
  • $0.10 per 1M tokens (legacy pricing)
  • Acceptable for many use cases

​
Filtering Models

​
By Capability

import requests

response = requests.get(
    "https://api.redpill.ai/v1/embeddings/models",
    headers={"Authorization": f"Bearer YOUR_API_KEY"}
)

models = response.json()["data"]

# Filter by dimension size
large_models = [m for m in models if m.get("max_output_length", 0) >= 2048]

# Filter by provider
openai_models = [m for m in models if m["id"].startswith("openai/")]

# Filter by pricing
cheap_models = [m for m in models if float(m["pricing"]["prompt"]) < 0.0000001]

# Find models supporting specific input
text_models = [m for m in models if "text" in m["input_modalities"]]

​
Comparison with Chat Models

AspectEmbedding ModelsChat Models
Endpoint/v1/embeddings/models/v1/models
OutputVector representationsText responses
Use CasesSearch, clustering, similarityConversation, generation, reasoning
PricingPer input token onlyPer input + output token
DimensionsFixed (configurable for some)N/A

​
Privacy & Security

TEE-Protected Embedding

All embedding generation flows through hardware-protected gateway

Confidential Queries

Your text/images processed in secure enclaves

No Data Retention

Embeddings not stored after generation

No Training Use

Your data never used to train models

​
Examples

​
List and Compare Models

import requests

def compare_embedding_models():
    response = requests.get(
        "https://api.redpill.ai/v1/embeddings/models",
        headers={"Authorization": f"Bearer YOUR_API_KEY"}
    )

    models = response.json()["data"]

    print("Model Comparison:\n")
    print(f"{'Model ID':<40} {'Dimensions':<12} {'Cost (per 1M tokens)'}")
    print("-" * 80)

    for model in models:
        model_id = model["id"]
        dimensions = model.get("max_output_length", "N/A")
        cost = float(model["pricing"]["prompt"]) * 1_000_000
        print(f"{model_id:<40} {str(dimensions):<12} ${cost:.4f}")

compare_embedding_models()

​
Find Cheapest Model

def find_cheapest_model(min_dimensions=768):
    response = requests.get(
        "https://api.redpill.ai/v1/embeddings/models",
        headers={"Authorization": f"Bearer YOUR_API_KEY"}
    )

    models = response.json()["data"]

    # Filter by minimum dimensions
    qualified = [
        m for m in models
        if m.get("max_output_length", 0) >= min_dimensions
    ]

    # Sort by price
    cheapest = min(qualified, key=lambda m: float(m["pricing"]["prompt"]))

    print(f"Cheapest model with {min_dimensions}+ dimensions:")
    print(f"  Model: {cheapest['id']}")
    print(f"  Dimensions: {cheapest['max_output_length']}")
    print(f"  Cost: ${float(cheapest['pricing']['prompt']) * 1_000_000:.4f} per 1M tokens")

    return cheapest["id"]

​
Error Handling

{
  "error": {
    "message": "Invalid API key",
    "type": "authentication_error",
    "code": 401
  }
}

​
Next Steps

Generate Embeddings

Create embeddings with these models

Multi-Modal Embeddings

Use image and video embeddings

Supported Models

Browse all available models

Pricing

View embedding costs