# Atlas AI Services Pricing (Agent Engine and Voyage AI) > Atlas AI spend has two metered parts: Voyage AI embedding and reranking (per token, or per pixel for multimodal models, with 200 million free tokens for most current models) billed to the Atlas organization that owns the API key, and Automated Embedding in Atlas Vector Search, which uses the same per-token rates and meters initial sync, inserts, updates and queries. Atlas Agent Engine (Agent Runtime and Agent Memory), announced 2026-09-29 in public preview, is consumption-priced and draws down existing Atlas commitments, but no unit rates are published. The pack draws on 6 sources (the Atlas Embedding and Reranking API and Automated Embedding billing pages, the Automated Embedding models page, the Voyage pricing page, a MongoDB press release and IT Brief coverage) and flags 2 disagreements about which reranker and which legacy embedding models receive free tokens. Parent: [Atlas Pricing Programs, Cost Governance and TCO](https://llms-explorer.com/tree/atlas-pricing-programs-cost-governance-and-tco/) · 10 facets · 85 facts · page: https://llms-explorer.com/tree/atlas-ai-services-pricing-agent-engine-and-voyage-ai/ ## Definitions - Atlas Embedding and Reranking API is pay-as-you-go: text models bill per token, multimodal models per pixel. Billing is set at organization level, and charges go to the Atlas account linked to the model API key. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Atlas Agent Engine was announced 2026-09-29 and is in public preview. Atlas Agent Runtime and Atlas Agent Memory are consumption-priced, and usage counts against existing Atlas commitments, so no new contract is needed. — [source](https://www.mongodb.com/de-de/company/newsroom/press-releases/mongodb-launches-atlas-agent-engine-to-put-ai-agents-in-production-without-a-new-stack) - MongoDB has launched Atlas Agent Engine, a new layer for running AI agents in production. The product is its latest effort to make the Atlas platform more central to enterprise AI deployments. — [source](https://itbrief.co.uk/story/mongodb-launches-atlas-agent-engine-for-ai-production) ## Structure and components - Automated Embedding uses Voyage AI's embedding models, which Atlas hosts and manages in the Data Plane in a multi-tenant environment. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) - Memory and retrieval are another part of the offering. Atlas Agent Engine uses Voyage AI embeddings and MongoDB's retrieval technology so agents can maintain context across tasks and interactions. — [source](https://itbrief.co.uk/story/mongodb-launches-atlas-agent-engine-for-ai-production) - Memory and retrieval built in. Without built-in memory, agents start every conversation from zero, and teams end up rebuilding memory infrastructure for every new agent. Atlas Agent Engine builds memory into the platform itself, using Voyage AI embeddings and MongoDB's native retrieval, so agents get more accurate while spending fewer tokens. — [source](https://www.mongodb.com/de-de/company/newsroom/press-releases/mongodb-launches-atlas-agent-engine-to-put-ai-agents-in-production-without-a-new-stack) - Built on Atlas, Atlas Agent Engine combines execution, memory and governance in a single layer. Customers can use the memory and governance components on their own or with MongoDB's runtime, while continuing to use existing AI models and frameworks. — [source](https://itbrief.co.uk/story/mongodb-launches-atlas-agent-engine-for-ai-production) ## How it works - Automated Embedding in Atlas Vector Search bills the same per-token model rates. Tokens are metered at index creation (initial sync), on document inserts and updates, and on every $vectorSearch query. The free allowance is 200M tokens per model, one-time, shared across all projects and clusters in the org, and it does not refresh. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/billing/) - Voyage model pricing is usage-based, with charges billed to the account linked to the API key used for access. — [source](https://docs.voyageai.com/pricing) - You can manage payments and billing for the Atlas Embedding and Reranking API through the billing and payments interface in the Atlas UI. Atlas sends billing charges to the Atlas account linked to your model API keys. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Embedding model pricing is usage-based, with charges billed to the account linked to the API key used for access. Usage is based on the number of tokens embedded during: — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/billing/) - Automated Embedding uses a token-based pricing model where you are charged based on the number of tokens processed by the embedding models. To learn more about: — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/billing/) - Pricing varies by model. For models that process text, charges apply per token in your text field and queries. To see the cost per 1,000 tokens for each supported embedding model, see Models for Automated Embedding. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/billing/) - The first-time index build rate limits restrict the maximum frequency and number of tokens at which embeddings are generated. For large workloads during the first-time index build (initial sync), Automated Embedding uses a separate inference mechanism that is not bound by standard rate limits. This mechanism is optimized for throughput to handle the initial index build, provides the following benefits: — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) - For text-based models, charges apply per token. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - For multimodal models that process images and videos, charges apply per pixel. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - You can manage payments and billing for Automated Embedding through the billing and payments interface in the Atlas UI. Atlas sends billing charges to the Atlas account linked to your cluster. Your billing profile for Atlas is the same as the billing profile for your Automated Embedding. To view or edit your billing profile, see Atlas Billing. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/billing/) - Model tokens are consumed during index operations (first time creation, inserts, update) and query operations. For the index operations, only the fields in the MongoDB document that are indexed as the autoEmbed type are used for embedding generation and incur a token usage. For the query operation, the query text provided is used for embedding generation and incurs a token usage. The cost for tokens for each model is as follows: — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) - You configure billing settings at the organization level, which apply to all projects within that organization. You can examine costs and billing for your model API keys per project in the Atlas UI. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) ## Parameters and configuration - Rate limits (RPM and TPM) are set at org level and can be lowered per project. Usage tier rises with cumulative spend: add a payment method for Tier 1, $100 for Tier 2, $1,000 for Tier 3. A linked child org takes the higher of its own and its parent's tier. Automated Embedding rate limits apply per cluster instead. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Automated Embedding rate limits, index insert and update, Model | RPM | TPM: voyage-4-large | 2,000 | 3,000,000; voyage-4 | 2,000 | 8,000,000; voyage-4-lite | 2,000 | 16,000,000; voyage-code-4 | 2,000 | 3,000,000; voyage-code-3 | 2,000 | 3,000,000. Query limits for M0 clusters without a payment method: 3 RPM and 2,000 TPM per model; paid clusters (M0 with payment method, Flex, Dedicated): 2,000 RPM with the same TPM as above. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) - Usage tiers, Usage Tier | Requirements to Advance: Free trial | Add a payment method to move to Usage Tier 1; Usage tier 1 | Spend at least $100 since account creation to move to Usage Tier 2; Usage tier 2 | Spend at least $1,000 since account creation to move to Usage Tier 3; Usage tier 3 | If you still need a higher rate limit, please request a rate limit increase. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - The Atlas Embedding and Reranking API enforces rate limits based on requests per minute (RPM) and tokens per minute (TPM). MongoDB sets limits at the organization level, but you can configure lower limits at the project level. Higher rate limit tiers are available based on payment history and spending. To learn more, see Manage Rate Limits. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Rate limits are restrictions on the frequency and number of tokens you can request from Automated Embedding within a specified period of time. MongoDB enforces rate limits on embedding generation to ensure fair usage across all users in the multi-tenant environment. Rate limits are based on the requests per minute (RPM) and tokens per minute (TPM). These rate limits apply at a MongoDB Cluster level and are shared between all the indexes on that cluster using Automated embedding. To request higher rate limits, contact your MongoDB account team or MongoDB support. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) - Rate limits are applied separately to queries, first-time index builds, and index update operations (document inserts and updates), providing traffic isolation. Indexing build operations are strictly isolated from the real-time query traffic. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) - To view billing information, you must have Organization Billing Viewer access or higher. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - To configure billing settings, such as the payment method and billing profile, you must have Organization Owner or Organization Billing Admin access. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - As your API usage and spending increase, you are automatically graduated to higher usage tiers with increased rate limits. Your organization is assigned to a usage tier based on cumulative spending since account creation. Once you qualify for a tier, you remain at that tier unless your organization is unlinked from a parent organization. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - When a child organization is linked to a parent organization, the child organization inherits the parent's usage tier if the parent's tier is higher than the child's current tier. To learn how to link organizations, see Cross-Organization Billing. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - When organization linking occurs, Atlas evaluates both the parent and child organization tiers and sets the child's tier to the maximum of the two. This ensures: — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - When an organization is unlinked from its parent, the child organization's tier is re-evaluated based on its own lifetime spend and payment method status. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) ## How-to and procedures - Cost visibility: Cost Explorer can filter by AI model, and Automated Embedding has its own Service filter value. Metering-derived costs refresh every 24 hours. Invoices break out usage per model and per project. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Your invoice shows usage broken down by embedding model (such as voyage-4, voyage-4-lite, voyage-4-large, voyage-code-4, and voyage-code-3), including the number of tokens processed and the cost per model. To learn more about supported models and their pricing, see Models for Automated Embedding. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/billing/) - To understand how your usage translates to costs, use the Cost Explorer in the Atlas UI and filter by Automated Embedding. MongoDB derives costs from metering data and updates them every 24 hours. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/billing/) - To filter by Automated Embedding, select Automated Embedding from the Service dropdown menu. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/billing/) - Tracks your usage of tokens and pixels on the Usage page. To learn more, see Monitor Usage. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Provides detailed cost breakdowns through the Cost Explorer. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Create an Index on Prepopulated Collections When Possible: The first-time index build (initial sync) benefits from a special inference mechanism that is not throughput bounded. Therefore, we recommend prepopulating your collection before creating an index. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) - Upgrade When Needed: If you consistently hit rate limits, consider adding a payment method for higher quotas, or reaching out to your MongoDB account team for higher rate limits. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) ## Examples and snippets - Multimodal image pricing examples, Image resolution | Number of pixels | Price per image | Price per 1K images: 200px x 200px | 40,000 | $0.00003 | $0.03; 1000px x 1000px | 1 million | $0.0006 | $0.60; 2000px x 2000px | 4 million | $0.0012 | $1.20; 4000px x 4000px | 16 million | $0.0012 | $1.20. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - If a child organization inherited Tier 3 from its parent but has only spent $50 on its own, unlinking will downgrade the child organization to the Free Tier (if no payment method is attached) or Tier 1 (if a payment method is attached). — [source](https://www.mongodb.com/docs/voyageai/management/billing/) ## Measurements and reference values - Voyage pricing page, older models (no free tokens), Model | Price per 1M tokens: voyage-context-3, voyage-code-3, voyage-3-large | $0.18; voyage-3.5 | $0.06; voyage-3.5-lite | $0.02; voyage-multilingual-2, voyage-large-2-instruct, voyage-large-2 | $0.12; voyage-01, voyage-lite-01, voyage-lite-01-instruct, voyage-02, voyage-lite-02-instruct, voyage-2 | $0.1; voyage-3 | $0.06; rerank-1, rerank-2, rerank-2.5 | $0.05; voyage-3-lite, rerank-lite-1, rerank-2-lite, rerank-2.5-lite | $0.02. — [source](https://docs.voyageai.com/pricing) *(older table lists voyage-code-3 and voyage-multilingual-2 without free tokens)* - Free tier is 200M tokens for most current models and 50M for voyage-finance-2, voyage-law-2 and voyage-code-2. Older models (voyage-3-large and voyage-code-3 at $0.18, voyage-3.5 at $0.06, voyage-3.5-lite at $0.02) get no free tokens. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Voyage pricing page, text embeddings, Model | Price per 1M tokens | Free tokens: voyage-4-large | $0.12 | 200 million; voyage-4 | $0.06 | 200 million; voyage-4-lite | $0.02 | 200 million; voyage-context-4 | $0.12 | 200 million; voyage-code-4 | $0.12 | 200 million; voyage-finance-2, voyage-law-2, voyage-code-2 | $0.12 | 50 million. — [source](https://docs.voyageai.com/pricing) - Voyage pricing page, rerankers, Model | Price per 1M tokens | Estimated price per request* | Number of free tokens: rerank-3 | $0.05 | $0.0025 | 200 million; rerank-3-lite | $0.02 | $0.001 | 200 million. *This price estimate assumes each request includes 100 documents and that the sum of the number of tokens in the query and the number of tokens in each document is 500. — [source](https://docs.voyageai.com/pricing) *(Voyage table lists rerank-3 and rerank-3-lite with 200 million free)* - Atlas Embedding and Reranking API, older text embedding models (free tokens are not offered), Model | Price per 1M tokens | Free tokens: voyage-3-large | $0.18 | 0; voyage-3.5 | $0.06 | 0; voyage-3.5-lite | $0.02 | 0; voyage-code-3 | $0.18 | 0. Older contextualized chunk embedding model: voyage-context-3 | $0.18 | 0. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) *(Atlas page lists voyage-code-3 as older, no free tokens)* - Atlas multimodal pricing, Model | Price per 1M tokens | Price per 1B pixels | Free tier: voyage-multimodal-3.5 | $0.12 | $0.60 | 200M tokens, 150B pixels. Older multimodal model, voyage-multimodal-3 | $0.12 | $0.60 | 0. Images are processed between 50,000 pixels (minimum) and 2 million pixels (maximum), with costs ranging from $0.00003 to $0.0012 per image. For pricing purposes, each video frame is considered an image. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Atlas rerankers, Model | Price per 1M tokens | Est. price per request* | Free tokens: rerank-3 | $0.05 | $0.0025 | 200 million; rerank-3-lite | $0.02 | $0.001 | 200 million. Older rerankers (free tokens are not offered): rerank-2.5 | $0.05 | $0.0025 | 0; rerank-2.5-lite | $0.02 | $0.001 | 0; rerank-2 | $0.05 | $0.0025 | 0; rerank-2-lite | $0.02 | $0.001 | 0. *Estimated price assumes 100 documents per request, with the sum of query tokens and tokens per document totaling 500. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) *(Atlas page: rerank-3 free; rerank-2.5 older, no free tokens)* - Current text embedding list prices per 1M tokens: voyage-4-large $0.12, voyage-4 $0.06, voyage-4-lite $0.02, voyage-code-4 $0.12. voyage-finance-2, voyage-law-2 and voyage-code-2 are $0.12. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Multimodal Voyage models (voyage-multimodal-3.5 and 3) cost $0.12 per 1M text tokens plus $0.60 per billion pixels, with 200M text tokens and 150B pixels free. — [source](https://docs.voyageai.com/pricing) - Rerank prices are $0.05 per 1M tokens for the standard reranker and $0.02 for the lite one. The Voyage pricing page names the models rerank-3 and rerank-3-lite, while search summaries cited rerank-2.5. The model name needs checking before it goes in a TCO model. — [source](https://docs.voyageai.com/pricing) *(rerank-3 per pricing table; rerank-2.5 per search summaries)* - Models for Automated Embedding, Embedding Model | Price Per 1M Tokens: voyage-4-lite | $0.02; voyage-4 (Recommended) | $0.06; voyage-4-large | $0.12; voyage-code-4 (Recommended for code) | $0.12; voyage-code-3, Legacy model specialized for code search and technical documentation. Use voyage-code-4 instead. | $0.18. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) *(Automated Embedding lists voyage-code-3 as legacy at $0.18)* - We charge for requests to the Voyage text embedding endpoint based on the number of tokens in the documents/queries. The first 200 million tokens for voyage-4-large, voyage-4, voyage-4-lite, voyage-context-4, and voyage-code-3, or the first 50 million tokens for voyage-multilingual-2, voyage-finance-2, voyage-law-2, and voyage-code-2, are free for every account. Subsequent usage is priced per token, as shown in the following table. — [source](https://docs.voyageai.com/pricing) *(Voyage prose lists voyage-code-3 and voyage-multilingual-2 as free; older table says no free tokens)* - Pricing for the Voyage reranker endpoint is based on the total number of processed tokens, calculated as “(the number of query tokens × the number of documents) + sum of the number of tokens in all documents.” The first 200 million tokens for rerank-2.5, rerank-2.5-lite, rerank-2, and rerank-2-lite are free for each account. Subsequent usage is priced per token, as shown in the following table. — [source](https://docs.voyageai.com/pricing) *(Voyage prose names rerank-2.5 and rerank-2 as free; its table lists rerank-3)* - Model pricing is usage-based, with charges billed to the Atlas account linked to the API key used for access. All models include a free tier. Get started with 200 million free tokens for most models, or 50 million tokens for specialized models. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Pricing is based on the number of tokens in your documents and queries. The free tier includes 200 million tokens for most models, and 50 million tokens for the following specialized models: voyage-finance-2, voyage-law-2, voyage-code-2. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Older Text Embedding ModelsThe following table shows the pricing for older text embedding models. Free tokens are not offered for these models. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - We charge for requests to the Voyage multimodal endpoint based on the number of tokens in the text and the number of pixels in images and videos. The first 200M text tokens and 150B pixels for voyage-multimodal-3.5 and voyage-multimodal-3 are free for every account. Subsequent usage is priced per token and per pixel, as shown in the following table. For pricing purposes, each video frame is considered an image. — [source](https://docs.voyageai.com/pricing) - Atlas Agent Engine gives teams a modular way to put agents into production. Retrieval is powered by MongoDB Voyage AI, whose embedding and reranking models rank among the top performers on RTEB, a benchmark built to reflect real enterprise retrieval instead of academic datasets. Customers can adopt the memory and governance layers independently or with the runtime, using existing models and frameworks they know. Atlas Agent Engine is available today in public preview, with consumption-based pricing for Atlas Agent Runtime and Atlas Agent Memory. Usage draws on customers' existing Atlas commitm — [source](https://www.mongodb.com/de-de/company/newsroom/press-releases/mongodb-launches-atlas-agent-engine-to-put-ai-agents-in-production-without-a-new-stack) - Pricing is based on text tokens and image pixels. The free tier includes 200 million text tokens and 150 billion pixels for multimodal models. Images are processed between 50,000 pixels (minimum) and 2 million pixels (maximum), with costs ranging from $0.00003 to $0.0012 per image. For pricing purposes, each video frame is considered an image. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Pricing is based on total processed tokens, calculated as (query tokens × number of documents) + sum of tokens in all documents. The free tier includes 200 million tokens for the latest reranker models. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) *(free tier names the latest rerankers; table lists rerank-3)* - For each model, Atlas includes a one-time allocation of 200 million free tokens at the organization level. The organization shares free tokens across all Atlas projects and clusters within the organization. Free tokens do not refresh. To learn more about free tokens and usage tiers, see Rate Limits. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/billing/) - For each model, MongoDB Vector Search includes a one-time allocation of 200 million free tokens. The free tokens are shared across all clusters in the deployment. Free tokens do not refresh. To learn more about free tokens and usage tiers, see Rate Limits. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/billing/) *(self-managed variant: shared across all clusters in the deployment)* - The product is available in public preview. Pricing for Atlas Agent Runtime and Atlas Agent Memory will be consumption-based and will count against existing Atlas commitments. — [source](https://itbrief.co.uk/story/mongodb-launches-atlas-agent-engine-for-ai-production) - Free tokens do not refresh. For example, when new models are released, the free tokens for legacy models do not get reset. You can continue to use the remaining free tokens for the legacy model. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) - The following table shows the pricing for older models. Please note that we do not offer free tokens for them. — [source](https://docs.voyageai.com/pricing) - Images with fewer than 50,000 pixels are upscaled, processed, and charged as a 50,000-pixel image. Images containing over 2 million pixels are downsampled and charged as 2 million-pixel images. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) - Images with fewer than 50,000 pixels will be upscaled, processed, and charged as a 50,000-pixel image. Images containing over 2 million pixels will be downsampled and charged as 2 million-pixel images. Therefore, the minimum and maximum costs per image are $0.00003 and $0.0012, respectively. The following table provides examples of image pricing. — [source](https://docs.voyageai.com/pricing) - Our Batch API provides a simple way to process multiple requests efficiently to our current models. It offers a 12-hour completion window and a 33% discount compared to our standard endpoints. — [source](https://docs.voyageai.com/pricing) - Free token credits do not apply to Batch API usage. All successfully processed Batch API tokens are billed at the applicable discounted batch rate and do not deduct from your free token balance. — [source](https://docs.voyageai.com/pricing) - Our Files API provides access to a lightweight file storage service to upload batch processing requests, download results, and manage related files. You can store an unlimited number of files, each retained for 30 days before automatic deletion. Storage is priced at $0.05 per GB per month. — [source](https://docs.voyageai.com/pricing) ## Problems, failure modes and limitations - Neither the MongoDB announcement nor press coverage publishes Agent Engine unit rates, billing units or free-tier terms. Agent Memory uses Voyage AI embeddings, so token-metered embedding cost is likely part of it. Memory and governance can be adopted without the runtime. — [source](https://www.mongodb.com/de-de/company/newsroom/press-releases/mongodb-launches-atlas-agent-engine-to-put-ai-agents-in-production-without-a-new-stack) - Legacy model specialized for code search and technical documentation. Use voyage-code-4 instead. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) *(legacy model)* - voyage-code-4 requires an Atlas deployment. On self-managed deployments, use voyage-code-3 for code search and technical documentation. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) - If the indexed text field is longer than the context window, the text is automatically truncated to the context window size for the model. If your query text exceeds this context window for the model, the $vectorSearch query fails with a context-limit-exceeded error. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) ## Facts and statements - Atlas Agent Engine is available today in public preview. New and existing Atlas customers can get started at agentengine.mongodb.com. — [source](https://www.mongodb.com/de-de/company/newsroom/press-releases/mongodb-launches-atlas-agent-engine-to-put-ai-agents-in-production-without-a-new-stack) - Governed by default. Most platforms handle identity, audit, guardrails, and cost controls as separate systems teams have to stitch together themselves. Atlas Agent Engine puts it all behind one control plane: every action is logged against a real identity, human or agent, and governed by policy that can't be quietly switched off. Governance is built in, not bolted on after launch. So when someone asks what an agent did and who authorized it, the answer takes seconds, not weeks. And because governance, memory, and retrieval run as one system instead of stitched-together services, there's less t — [source](https://www.mongodb.com/de-de/company/newsroom/press-releases/mongodb-launches-atlas-agent-engine-to-put-ai-agents-in-production-without-a-new-stack) *(cost controls named, no rates published)* ## Disagreements - Rerank prices are $0.05 per 1M tokens for the standard reranker and $0.02 for the lite one. The Voyage pricing page names the models rerank-3 and rerank-3-lite, while search summaries cited rerank-2.5. The model name needs checking before it goes in a TCO model. — [source](https://docs.voyageai.com/pricing) *(rerank-3 per pricing table; rerank-2.5 per search summaries)* - Pricing for the Voyage reranker endpoint is based on the total number of processed tokens, calculated as “(the number of query tokens × the number of documents) + sum of the number of tokens in all documents.” The first 200 million tokens for rerank-2.5, rerank-2.5-lite, rerank-2, and rerank-2-lite are free for each account. Subsequent usage is priced per token, as shown in the following table. — [source](https://docs.voyageai.com/pricing) *(Voyage prose names rerank-2.5 and rerank-2 as free; its table lists rerank-3)* - Atlas rerankers, Model | Price per 1M tokens | Est. price per request* | Free tokens: rerank-3 | $0.05 | $0.0025 | 200 million; rerank-3-lite | $0.02 | $0.001 | 200 million. Older rerankers (free tokens are not offered): rerank-2.5 | $0.05 | $0.0025 | 0; rerank-2.5-lite | $0.02 | $0.001 | 0; rerank-2 | $0.05 | $0.0025 | 0; rerank-2-lite | $0.02 | $0.001 | 0. *Estimated price assumes 100 documents per request, with the sum of query tokens and tokens per document totaling 500. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) *(Atlas page: rerank-3 free; rerank-2.5 older, no free tokens)* - Voyage pricing page, rerankers, Model | Price per 1M tokens | Estimated price per request* | Number of free tokens: rerank-3 | $0.05 | $0.0025 | 200 million; rerank-3-lite | $0.02 | $0.001 | 200 million. *This price estimate assumes each request includes 100 documents and that the sum of the number of tokens in the query and the number of tokens in each document is 500. — [source](https://docs.voyageai.com/pricing) *(Voyage table lists rerank-3 and rerank-3-lite with 200 million free)* - We charge for requests to the Voyage text embedding endpoint based on the number of tokens in the documents/queries. The first 200 million tokens for voyage-4-large, voyage-4, voyage-4-lite, voyage-context-4, and voyage-code-3, or the first 50 million tokens for voyage-multilingual-2, voyage-finance-2, voyage-law-2, and voyage-code-2, are free for every account. Subsequent usage is priced per token, as shown in the following table. — [source](https://docs.voyageai.com/pricing) *(Voyage prose lists voyage-code-3 and voyage-multilingual-2 as free; older table says no free tokens)* - Atlas Embedding and Reranking API, older text embedding models (free tokens are not offered), Model | Price per 1M tokens | Free tokens: voyage-3-large | $0.18 | 0; voyage-3.5 | $0.06 | 0; voyage-3.5-lite | $0.02 | 0; voyage-code-3 | $0.18 | 0. Older contextualized chunk embedding model: voyage-context-3 | $0.18 | 0. — [source](https://www.mongodb.com/docs/voyageai/management/billing/) *(Atlas page lists voyage-code-3 as older, no free tokens)* - Voyage pricing page, older models (no free tokens), Model | Price per 1M tokens: voyage-context-3, voyage-code-3, voyage-3-large | $0.18; voyage-3.5 | $0.06; voyage-3.5-lite | $0.02; voyage-multilingual-2, voyage-large-2-instruct, voyage-large-2 | $0.12; voyage-01, voyage-lite-01, voyage-lite-01-instruct, voyage-02, voyage-lite-02-instruct, voyage-2 | $0.1; voyage-3 | $0.06; rerank-1, rerank-2, rerank-2.5 | $0.05; voyage-3-lite, rerank-lite-1, rerank-2-lite, rerank-2.5-lite | $0.02. — [source](https://docs.voyageai.com/pricing) *(older table lists voyage-code-3 and voyage-multilingual-2 without free tokens)* - Models for Automated Embedding, Embedding Model | Price Per 1M Tokens: voyage-4-lite | $0.02; voyage-4 (Recommended) | $0.06; voyage-4-large | $0.12; voyage-code-4 (Recommended for code) | $0.12; voyage-code-3, Legacy model specialized for code search and technical documentation. Use voyage-code-4 instead. | $0.18. — [source](https://www.mongodb.com/docs/vector-search/crud-embeddings/automated-embedding/models/) *(Automated Embedding lists voyage-code-3 as legacy at $0.18)* ## Related concepts - Voyage AI — is a related of Atlas AI Services Pricing (Agent Engine and Voyage AI) - free tokens — is a part of Atlas AI Services Pricing (Agent Engine and Voyage AI) - Automated Embedding — is a part of Atlas AI Services Pricing (Agent Engine and Voyage AI) - Atlas Agent Engine — is a part of Atlas AI Services Pricing (Agent Engine and Voyage AI) - per-token pricing — is a part of Atlas AI Services Pricing (Agent Engine and Voyage AI) - usage tiers and rate limits — is a part of Atlas AI Services Pricing (Agent Engine and Voyage AI) - Voyage AI embedding models — is a part of Atlas AI Services Pricing (Agent Engine and Voyage AI) - multimodal pricing — is a part of Atlas AI Services Pricing (Agent Engine and Voyage AI) - cost visibility — is a part of Atlas AI Services Pricing (Agent Engine and Voyage AI) - older models — is a contrast of Atlas AI Services Pricing (Agent Engine and Voyage AI) - reranker pricing — is a part of Atlas AI Services Pricing (Agent Engine and Voyage AI) - billing responsibility — is a measure of Atlas AI Services Pricing (Agent Engine and Voyage AI) - Agent Engine consumption pricing — is a measure of Atlas AI Services Pricing (Agent Engine and Voyage AI) - Batch and Files API — is a part of Atlas AI Services Pricing (Agent Engine and Voyage AI)