tools
Mixedbread Embed
Legacy direct embedding APIs and models from Mixedbread; new integrations are directed to Mixedbread Search Stores.
In inglese
Mixedbread Embed was a set of direct embedding APIs and open-source embedding models for creating vector representations of text and other content. Listed model families include mxbai-embed-large-v1, mxbai-embed-2d-large-v1, and mxbai-embed-xsmall-v1.
Mixedbread now labels the embedding documentation as legacy and recommends Stores for new managed-search integrations. Stores handle parsing, indexing, retrieval, and reranking, with usage-based pricing and optional paid plans.
Features
- Direct embedding APIs for generating vector representations
- Embedding model family mxbai-embed-large-v1
- Embedding model family mxbai-embed-2d-large-v1
- Embedding model family mxbai-embed-xsmall-v1
- Supports multilingual and multimodal search through current Stores
- Current workflow handles parsing, indexing, retrieval, and reranking
- API access for integrating search into applications
Use cases
- Generate embeddings for semantic search systems
- Build retrieval-augmented generation pipelines
- Migrate legacy embedding integrations to Mixedbread Stores
- Search documents, images, code, and video through a managed API
- Add ranked retrieval to applications and AI agents
Pros
Cons
Latest updates
- Tag Stores for Easy Tracking
Stores can be tagged for tracking and filtering, and copied Stores inherit their source's tags.
- Copy Store for Testing and Branching
You can copy a Store's files into a new Store without reprocessing; the copy inherits the source configuration by default.
- Structured Output for Toast 1 (Toast 1)
The Responses and Chat Completions APIs now support structured outputs for Toast 1.
- MCP Tools in the Responses API (Toast 1)
The Responses API can use external MCP servers and tools with Toast 1.
- Context Management for Toast 1 (Toast 1)
Responses and Chat Completions requests can enable context management to prune stale tool results.
Capabilities
- Knows the whole codebase — “Upload PDFs, images, documents, code, or video in any format and instantly make it searchable with natural language queries.” source
- Command line — “We built our CLI to streamline Store operations for developers.” source
- Self-hosted — “Bring Your Own Cloud runs Mixedbread inside your own cloud account, keeping data and compute there.” source
- No code retained — “Bring Your Own Bucket keeps your content in object storage you control, such as S3, while Mixedbread indexes and searches it without retaining it.” source
- API — “Mixedbread is an API for integrating fast, multimodal search into your applications, agents, and AI systems.” source
- Runs models for you — “They are separate from the LLM tokens Toast 1 uses for inference.” source
- Search over your data — “Create a store (your search index) and upload any file format — PDFs, images, documents, code, videos.” source
Get it
Pricing
- Starting price
- $20/mo
- Prices checked
- 2026-09-25
Starter
Free
- $5 in one-time credits
- 3 workspace users
- 10 stores
- 100 requests / minute
- Community Slack support
Scale
- $20 / month
- $20 of credits included every month
- Unlimited workspace users
- 10,000 stores
- 1,200 queries/min, 360 ingestion/min
- Automatic backups
- Priority Slack support (same-day SLA)
Enterprise
Price on request
- Volume-based discounts
- Unlimited workspace users
- Unlimited stores
- Custom rate limits & SLA
- Automatic backups, point-in-time recovery
- Dedicated support team