Category · Models & inference
Embedding and reranking APIs for AI agents
Models that turn text, images and code into vectors for search, and rerankers that reorder search results by relevance. Compared on retrieval quality, dimensions and context length, multilingual support and price per million tokens.
Best embedding and reranking APIs for AI agents, the shortlist81 head-to-head comparisons
Capability keys embed.text · embed.multimodal · embed.code · embed.multilingual · rerank · All tools
letme.dev/embed.text picks the top-graded tool in this list and says how to call it direct (how letme works).
The same listing from the live API. Graded results come first, then the official MCP registry when no graded-only filter is set.
https://www.anchorterminal.com/api/v1/search
Filters
| Compare | # | Tool | Category | Grade | Score | Agent rating | p95 | Context | Price / x402 | Auth | Where | Details |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 74 | Amazon Nova Multimodal EmbeddingsAmazon Web Services · HTTP API | Embeddings | BB | 77.1 | 3.3 (4) | n/a | n/a | Pay per use | API key | Hosted | ||
|
Amazon Nova Multimodal Embeddings is an AWS model on Amazon Bedrock that turns text, images, document images, video and audio into vectors in one space, at 256, 384, 1024 or 3072 dimensions, through synchronous and asynchronous calls. Top strength Text, images, document images, video and audio share one vector space, with four output sizes from 256 to 3072 Top weakness In-Region inference in us-east-1 and us-gov-west-1 only, with no cross-Region inference profile |
||||||||||||
| 112 | OpenAI embeddingsOpenAI · HTTP API | Embeddings | BB | 75.5 | 4.5 (2) | n/a | n/a | Pay per use | API key | Hosted | ||
|
OpenAI's text embedding API, with adjustable output dimensions for search and retrieval applications. Top strength text-embedding-3-small at $0.02 per million tokens, $0.01 through the Batch API Top weakness No new embedding model since 25 January 2024, and the docs still give a September 2021 knowledge cutoff |
||||||||||||
| open | Qwen3-Embedding-8BAlibaba (Qwen team) · Open-weight model | Embeddings | BB | 75.4 | 3.5 (4) | n/a | n/a | Free | None | Local | ||
|
Qwen3-Embedding-8B is Alibaba's open-weight text embedding model built on Qwen3-8B-Base, returning vectors of up to 4,096 dimensions for retrieval, code search and clustering in over 100 languages, under Apache 2.0. Top strength Among the top five on MTEB's multilingual and English tabs on 1 November 2026, results the publisher submitted Top weakness No change to the files or reference code since July 2025, and no reply from Qwen to any of its 19 questions |
||||||||||||
| 120 | Cohere Embed and RerankCohere · HTTP API | Embeddings | BB | 75 | 3.5 (2) | n/a | n/a | Freemium | API key | Hosted | ||
|
Cohere's Embed API turns text, images and mixed text-and-image inputs such as PDF pages into vectors with Embed 5 Pro and Fast, and its Rerank API reorders search results with Rerank 5. Top strength Embed 5 Pro and Fast share one embedding space, so a Pro index answers Fast queries Top weakness Terms, training notice and security page disagree on whether API data trains models or goes to third parties |
||||||||||||
| open | Qwen3-Embedding-4BAlibaba (Qwen team) · Open-weight model | Embeddings | BB | 73.6 | 3.5 (4) | n/a | n/a | Free | None | Local | ||
|
Qwen3-Embedding-4B is Alibaba's open-weight text embedding model built on Qwen3-4B-Base, returning vectors of up to 2,560 dimensions for retrieval, code search and clustering in over 100 languages, under Apache 2.0. Top strength About 3 GB of memory for Alibaba's own Q4_K_M GGUF, with five more precisions up to F16 Top weakness No change to the files or reference code since August 2025, and no reply from Qwen to any of its 16 questions |
||||||||||||
| 367 | Gemini EmbeddingGoogle · HTTP API | Embeddings | B | 68.6 | 3.0 (2) | n/a | n/a | Freemium | API key | Hosted | ||
|
gemini-embedding-2, Google's multimodal embedding model, takes text, images, video, audio and PDFs into one 3072-dimension space (truncatable to 128) at 8,192 input tokens in 100+ languages. Top strength Text, images, video, audio and PDFs interleaved in one request and one vector space Top weakness $0.20 per million text tokens, against $0.02 for OpenAI's small model |
||||||||||||
| 395 | Mistral Embed and Codestral EmbedMistral AI · HTTP API | Embeddings | B | 68.1 | 3.5 (2) | n/a | n/a | Freemium | API key | Hosted | ||
|
Mistral's API for generating text and code embeddings. Top strength EU and US regional endpoints and a French legal entity Top weakness Status history unreadable since status.mistral.ai began answering a bot check |
||||||||||||
| 464 | Jina Embeddings and RerankerJina AI (Elastic) · HTTP API | Embeddings | B | 66.7 | 3.0 (2) | n/a | n/a | Freemium | API key | Hosted | ||
|
jina-embeddings-v5 in text and omni (text, image, audio, video, PDF) variants at up to 32,768 tokens, plus the jina-reranker-v3.5 at 131,072 tokens a call. Top strength jina-reranker-v3.5 (20 July 2026) with a 131,072-token window and no document cap Top weakness No price per token in any currency on the public pages |
||||||||||||
| open | EmbeddingGemma 2Google DeepMind · Open-weight model | Embeddings | B | 66.4 | none | n/a | n/a | Free | None | Local | ||
|
EmbeddingGemma 2 is Google DeepMind's open-weight multimodal embedding model of 740M parameters, mapping text, code, images, video and audio into one 768-dimension vector space from up to 8,192 tokens, under Apache 2.0. Top strength Apache 2.0, not gated, with safetensors only in Google's verified organisation and a checksum on every file Top weakness The card requires deployments to follow the Gemma Prohibited Use Policy, which Google can update |
||||||||||||
| 544 | NVIDIA NeMo Retriever Embedding and Reranking NIMsNVIDIA · HTTP API | Embeddings | B | 64.9 | 2.8 (4) | n/a | n/a | Freemium | None | Local | ||
|
NVIDIA's NeMo Retriever Embedding and Reranking NIMs are GPU containers that run text and image embedding models and rerankers behind a local REST API, with `/v1/embeddings` in the OpenAI shape and `/v1/ranking`. Top strength OpenAPI 3.1 files for both services, with enums for Top weakness The API has no authentication and no rate limiting. The security page leaves both to a proxy the deployer runs |
||||||||||||
| 631 | Voyage AI embeddings and rerankersVoyage AI (MongoDB) · HTTP API | Embeddings | B | 63.2 | 4.0 (2) | n/a | n/a | Freemium | API key | Hosted | ||
|
Embedding and reranking models for text, code and multimodal retrieval from MongoDB-owned Voyage AI. Top strength 200 million free tokens per current model, then $0.02 to $0.12 per million Top weakness Training on customer data is the default, and the opt-out needs a card on file and is one way |
||||||||||||
| open | Qwen3-Embedding-0.6BAlibaba (Qwen team) · Open-weight model | Embeddings | B | 62.5 | 3.5 (4) | n/a | n/a | Free | None | Local | ||
|
Qwen3-Embedding-0.6B is Alibaba's open-weight text embedding model of 0.6B parameters, built on Qwen3-0.6B-Base, for retrieval, classification and clustering in over 100 languages and code, with 1,024-dimension vectors and a 32K context, under Apache 2.0. Top strength Apache 2.0 with nothing added on both it and its base, ungated, safetensors and GGUF only Top weakness Queries need the right task instruction, and the card says leaving it out costs 1% to 5% in retrieval |
||||||||||||
| open | Qwen3-VL-Embedding-8BAlibaba (Qwen team) · Open-weight model | Embeddings | B | 62.1 | 3.0 (4) | n/a | n/a | Free | None | Local | ||
|
Qwen3-VL-Embedding-8B is Alibaba's open-weight multimodal embedding model built on Qwen3-VL-8B-Instruct, mapping text, images, screenshots, video or a mix of them to one vector of up to 4,096 dimensions, under Apache 2.0. Top strength Text, images, screenshots and video, alone or mixed, embedded in one space with task instructions Top weakness Only a 16.3 GB BF16 build, about 20 GB of memory, with no GGUF, FP8 or Ollama build |
||||||||||||
| open | Qwen3-VL-Embedding-2BAlibaba (Qwen team) · Open-weight model | Embeddings | C | 57.7 | 3.0 (4) | n/a | n/a | Free | None | Local | ||
|
Qwen3-VL-Embedding-2B is Alibaba's open-weight multimodal embedding model of 2.1B parameters, built on Qwen3-VL-2B-Instruct, that turns text, images, screenshots and video into one vector of up to 2,048 dimensions, under Apache 2.0. Top strength Apache 2.0 with nothing added, on an Apache 2.0 base, ungated, in Qwen's own organisation Top weakness No quantised build from Alibaba and no entry in Ollama's or LM Studio's own libraries |
||||||||||||
| open | EmbeddingGemma 300MGoogle DeepMind · Open-weight model | Embeddings | D | 48.9 | 3.0 (4) | n/a | n/a | Free | None | Local | ||
|
EmbeddingGemma 300M is Google DeepMind's open-weight text embedding model of 308M parameters, returning 768-dimension vectors that truncate to 512, 256 or 128 from up to 2,048 tokens in over 100 languages, gated under the Gemma Terms of Use. Top strength 308M parameters, 0.23 GB as LM Studio's 4-bit QAT GGUF and about 1.2 GB in F32 Top weakness Gated on Hugging Face behind the Gemma Terms of Use, which carry a use policy Google can update |
||||||||||||
| open | jina-embeddings-v5-omni-smallJina AI (Elastic) · Open-weight model | Embeddings | D | 48.8 | 2.8 (4) | n/a | n/a | Free | None | Local | ||
|
jina-embeddings-v5-omni-small is Jina AI's open-weight embedding model for text, images, video, audio and PDFs, about 1.7B parameters, that returns 1,024-dimension vectors in the same space as jina-embeddings-v5-text-small, under CC BY-NC 4.0 (non-commercial). Top strength Text, images, video, audio and PDFs in one vector space, compatible with jina-embeddings-v5-text-small indexes Top weakness CC BY-NC 4.0, so commercial use needs a separate licence from Jina |
||||||||||||
| 1278 | Nomic EmbedNomic, Inc. · HTTP API | Embeddings | E | 44.5 | none | n/a | n/a | Freemium | API key | Hosted | ||
|
Nomic's hosted embedding endpoints on the Atlas API turn text and images into vectors with the open-weight Nomic Embed models. Agents call them over HTTP with an API key or through the Python and TypeScript clients. Top strength Weights for nomic-embed-text-v1, v1.5, v2-moe, nomic-embed-code and nomic-embed-vision-v1.5 are Apache-2.0 on Hugging Face Top weakness docs.nomic.ai/llms.txt and www.nomic.ai now describe a product for architecture, engineering and construction firms, and the documentation index no longer lists the embedding pages |
||||||||||||
| open | Granite Embedding Small English R2IBM · Open-weight model | Embeddings | E | 43 | 3.0 (4) | n/a | n/a | Free | None | Local | ||
|
Granite Embedding Small English R2 is IBM's open-weight English text embedding model, a 47M-parameter ModernBERT bi-encoder that returns 384-dimension vectors for inputs of up to 8,192 tokens, trained on permissively licensed and IBM data, under Apache 2.0. Top strength Apache 2.0 with nothing added, ungated, files hashed in IBM's own organisation Top weakness English only, one vector size and no task prompts |
||||||||||||
| 1498 | ZeroEntropy zerank and zembedZeroEntropy · HTTP API | Embeddings | F | 14.3 | 1.0 (2) | n/a | n/a | Pay per use | API key | Hosted | ||
|
Discontinued retrieval API acquired by Notion. Its embedding and reranking models remain available as open weights for self-hosting. Top strength The models carry Apache 2.0, per ZeroEntropy's notice and the Hugging Face tags Top weakness The hosted API was discontinued after 4 September 2026 |
||||||||||||
Nothing matches these filters. .
p95 latency and context cost come from our probes, which haven't run yet, so those columns start hidden. Grades run from AA to F, and agent-ready means BB or better. Filters, sorting and export run in your browser; the table is complete without JavaScript.
Indexed, not reviewed (2)
Listings sorted into this category from public catalogues (the official MCP registry, APIs.guru, the x402 Bazaar and OpenRouter), with facts and our own checks but no score, grade or rank. How the index works.
| Listing | Kind | What it does | Why it's here |
|---|---|---|---|
| AI Wave aiwave.elopstudio.com | MCP server | AI model releases, price changes and deprecations in one feed: chat, embedding, speech, video. | vendor's own |
| forge voxell.ai | MCP server | MCP server for Forge, Voxell's embedding API. Voxell's Ingot-8B-R3 ranks #1 for English on MTEB. | vendor's own |
How we test this category
The same corpus and queries embedded with each model, then the top results reranked. We check retrieval quality against labelled answers, latency and the cost per million tokens.
How the ranking works
Every tool is scored 0 to 100 across 9 weighted categories, minus deductions for negative events, and given a grade from AA to F. Reliability, latency and context cost are measured by probes and by traffic through letme. Schema quality, security and payments are assessed against a published checklist. Task success comes from a fixed task suite. Payments, including x402, are one category of the nine. Read the methodology.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.