Which embedding model should I pick?

text-embedding-3-large is a strong default in the OpenAI ecosystem. Voyage 3 is excellent for long-document retrieval. text-embedding-3-small is the cheapest option for high-volume workloads.

What is the “intelligence index”?

A 0–100 composite of public benchmarks (MMLU, MMLU Pro, GPQA, MATH, HumanEval). Higher is better.

How often is this updated?

Monthly. Each model has a lastUpdated stamp. Open an issue if you spot something stale.

AI Models · Embeddings

Best Embeddings AI Models

Embedding models for search, RAG, clustering, and semantic similarity.

Models

Providers

Best Embedding Models (2026): OpenAI, Voyage, Gemini & More

4 models matched. Click any column to sort.

ProviderModel typeOpen sourceContextPriceRelease year

Capabilities

					Notes
Gemini Embedding 001	Google	Proprietary	$0.15	2k	Google’s general-purpose embedding model for Vertex AI and Gemini API.
Voyage 3	Voyage AI	Proprietary	$0.06	32k	Retrieval-focused embeddings with a large context window for long docs.
text-embedding-3-large	OpenAI	Proprietary	$0.13	8k	OpenAI’s flagship embedding model for RAG, search and clustering.
text-embedding-3-small	OpenAI	Proprietary	$0.02	8k	Cheap, fast OpenAI embeddings for high-volume retrieval workloads.

Showing 4 of 4 models. Click any column header to sort. Prices are USD per 1M tokens unless noted otherwise. Estimates marked with *.

Browse AI Models by category

Drill into a slice of the catalog — reasoning, coding, image, video, speech, embeddings, agents, open weights, and more.

By Use case

Frontier

The most capable models from each major lab.

Reasoning Models

Long chain-of-thought models built for hard math, code, and planning.

Coding Models

Models specialised for software engineering and code generation.

Agents

Models strong at tool use, multi-step workflows, and agentic systems.

Fast & Cheap

Production-grade workhorses with the best speed and cost.

Multimodal

Models that natively understand text, images, and beyond.

By Modality

Image Models

Generate and edit still images from text or reference images.

Video Models

Generate short video clips from text or image prompts.

Audio Models

Models that understand or generate audio beyond plain speech TTS.

Speech Models

Text-to-speech and voice models for natural spoken audio.

By License

Open Source

Models with publicly-released weights you can self-host or fine-tune.

Open Weights

Downloadable weights for self-hosting, fine-tuning, and private deployment.

Proprietary

Closed-weights API-only models from major labs.

By Audience

Enterprise

Production-ready models for teams that need reliability, security, and scale.

Consumer

Models popular in consumer apps, chat products, and creative tools.

By Provider

OpenAI

All models from OpenAI — GPT, o-series and beyond.

Anthropic

Claude family of models from Anthropic.

Google

Gemini family of models from Google DeepMind.

Mistral

Models from Mistral AI — EU-hosted, multilingual.

DeepSeek

Frontier-class open-weights models from DeepSeek.

xAI

Grok models from xAI.

Alibaba

Qwen family of open-weights models from Alibaba.

Voyage AI

Embedding models from Voyage AI.

Midjourney

Midjourney text-to-image models.

Black Forest Labs

FLUX image-generation models from Black Forest Labs.

Stability AI

Stable Diffusion image models from Stability AI.

Runway

Gen-series text-to-video models from Runway.

Kuaishou

Kling text-to-video models from Kuaishou.

ElevenLabs

Text-to-speech voice models from ElevenLabs.

Cartesia

Sonic low-latency text-to-speech models from Cartesia.

Frequently asked questions

Embedding models turn text into vectors for semantic search, RAG, clustering, and recommendations. Unlike chat models, they don’t generate answers — they power retrieval that chat models then reason over.

Explore the full catalog

See every AI model in one place — intelligence, speed and price on a single sortable table.

All AI models LLM pricing calculator