PRICING

Only pay for what you use.

Powabase is the Postgres backend for AI apps, with RAG and agents built in. Start free. Every paid plan comes with a monthly credit balance equal to the subscription, which you can spend on compute hours, per-call costs, MAU overages, and storage. Unused credits roll over month to month. The bigger the plan, the cheaper each unit.

Bring your own LLM keys for OpenAI, Anthropic, Google, and OpenRouter. API keys are stored per-project and encrypted at rest. Or skip keys entirely and pay for inference from your Powabase credits.

TWO WAYS TO RUN IT

Self-host it for free, or let us run it.

OPEN SOURCE

Self-hosted

FreeApache-2.0, forever

Run the full single-project stack on your own infrastructure with one docker compose up. Yours to operate, with no limits and no account.

  • Full stack: Postgres, auth, storage, REST, and the AI service
  • One docker compose up, no control plane
  • Bring your own LLM keys
  • Community support on Discord
MANAGED

Powabase Cloud

From $0usage-based

We provision, scale, back up, and update every project. Start free; paid plans add prepaid credits and cheaper usage. See the plans below.

  • Provisioned in seconds, fully managed
  • Per-project isolation, backups, and scaling
  • Free tier, then Self Serve / Scale / Enterprise
  • Email and priority support

POWABASE CLOUD PLANS

Managed hosting on Powabase Cloud. Prefer to run it yourself? The self-hosted edition above is free and open source.

Free

$0forever

All platform features. Sign up, start a project, ship.

  • $10 in free credits on sign-up
  • Per-project Postgres + pgvector
  • Auth, Storage, Realtime
  • RAG pipeline with OCR + four indexing strategies
  • Agents, multi-agent orchestrations, workflows
  • Vector / BM25 / hybrid / tree retrieval + reranking
  • Built-in tools + custom HTTP + MCP servers
  • Bring your own LLM keys
  • Pay-as-you-go, with no monthly minimum or commitment
  • Inactive projects suspend after 7 days, delete after 30

Self Serve

$25/month

For solo devs and small teams running steady workloads.

$25 in monthly credits to spend on anything

  • Pick any compute tier (Sandbox → Foundry)
  • Up to 25% cheaper per-call costs vs. Free
  • 15% cheaper per-hour compute vs. Free
  • Lower overage rates on EBS, S3, egress, requests
  • Email support

Scale

$300/month

For teams running production AI workloads at higher throughput.

$300 in monthly credits to spend on anything

  • Up to 50% cheaper per-call costs vs. Free
  • 20% cheaper per-hour compute vs. Free
  • Lowest overage rates on EBS, S3, egress, requests
  • Priority live support

Enterprise

Custom

Production-grade scale, procurement-ready terms, and a deployment model that fits your security posture.

Managed on Powabase Cloud or private hosting in your VPC / on-prem

  • Free MVP build for select annual commitments
  • BYO Cloud or data center
  • Regional data residency (US, EU); air-gapped supported
  • SOC 2, ISO 27001, DPA, SLA
  • Cybersecurity insurance coverage
  • SSO (SAML / OIDC), audit logs, RBAC
  • Priority support with SLAs + named contact
  • Dedicated solutions engineer

See per-call and per-tier rates below — paid plans pay the cheaper columns.

INFRASTRUCTURE

Compute, storage, and auth.

Every project runs on its own isolated stack. Pick a tier by workload size. Compute is billed per hour, with bundled storage, egress, and request quotas. Auth metering (MAU, SSO, Phone MFA) is billed separately at flat rates.

Per-hour rates step down as you upgrade plans (Free → Self Serve → Scale).

Sandbox

Postgres

0.5 vCPU / 500 MB

AI Runtime

0.5 vCPU / 500 MB

Free

$0.0346

Self Serve

$0.0296

Scale

$0.0272

Default for Free plan projects

Builder

Postgres

0.5 vCPU / 1 GiB

AI Runtime

1 vCPU / 1.5 GiB

Free

$0.0470

Self Serve

$0.0403

Scale

$0.0370

Solo dev's first agentic feature

Workshop

Postgres

1 vCPU / 2 GiB

AI Runtime

1.5 vCPU / 3.5 GiB

Free

$0.0932

Self Serve

$0.0799

Scale

$0.0733

Team's first deployed AI app

Studio

Postgres

2 vCPU / 4 GiB

AI Runtime

3.5 vCPU / 8.5 GiB

Free

$0.2030

Self Serve

$0.1740

Scale

$0.1595

Multi-agent workloads

Foundry

Dedicated QoS

Postgres

4 vCPU / 8 GiB

AI Runtime

7 vCPU / 9 GiB

Free

$0.4032

Self Serve

$0.3456

Scale

$0.3168

High-throughput orchestrations

¹ AI runtime covers the API gateway, RAG pipeline, agent runtime, workflow engine, and sidecars (auth, storage, realtime, pg-meta, PostgREST). Isolated per project.

BUNDLED PER PROJECT

· Included with every paid tier
  • Sandbox

    EBS storage
    2 GB
    S3 storage
    1 GB
    Internet egress
    5 GB/mo
    S3 PUT
    10k/mo
    S3 GET
    50k/mo
  • Builder

    EBS storage
    8 GB
    S3 storage
    5 GB
    Internet egress
    20 GB/mo
    S3 PUT
    50k/mo
    S3 GET
    250k/mo
  • Workshop

    EBS storage
    8 GB
    S3 storage
    10 GB
    Internet egress
    40 GB/mo
    S3 PUT
    100k/mo
    S3 GET
    500k/mo
  • Studio

    EBS storage
    20 GB
    S3 storage
    30 GB
    Internet egress
    100 GB/mo
    S3 PUT
    250k/mo
    S3 GET
    1.5M/mo
  • Foundry

    EBS storage
    50 GB
    S3 storage
    75 GB
    Internet egress
    200 GB/mo
    S3 PUT
    750k/mo
    S3 GET
    3.5M/mo
See overage rates

Pass-through pricing. You're only charged for usage above your bundle.

  • EBS gp3 storage

    $/GB-mo

    Free

    $0.134

    Self Serve

    $0.115

    Scale

    $0.106

  • EBS gp3 IOPS overage

    $/IOPS-mo

    Free

    $0.007

    Self Serve

    $0.006

    Scale

    $0.0055

  • EBS gp3 throughput overage

    $/MB/s-mo

    Free

    $0.056

    Self Serve

    $0.048

    Scale

    $0.044

  • NAT Gateway data

    $/GB

    Free

    $0.063

    Self Serve

    $0.054

    Scale

    $0.0495

  • S3 storage

    $/GB-mo

    Free

    $0.0322

    Self Serve

    $0.0276

    Scale

    $0.0253

  • S3 PUT/COPY/POST/LIST

    $/1k req

    Free

    $0.0077

    Self Serve

    $0.0066

    Scale

    $0.00605

  • S3 GET/SELECT

    $/1k req

    Free

    $0.000616

    Self Serve

    $0.000528

    Scale

    $0.000484

  • Internet egress

    $/GB

    Free

    $0.126

    Self Serve

    $0.108

    Scale

    $0.099

AUTH METERING

Flat rates across all tiers, with no markup and no tier discount.

  • · Monthly active users authenticated via email, password, OAuth, or magic link.
  • · Enterprise SSO (SAML / OIDC) users. Not available on Free or Builder.
  • · Phone-based MFA. SMS billed at carrier list price + 10% with a default $50/mo spend cap.
  • Sandbox

    Regular MAU

    Included
    50,000
    Overage
    —

    SSO MAU

    Included
    —
    Overage
    not available

    Phone MFA + SMS

    MFA
    not available
    Per-SMS
    —
  • Builder

    Regular MAU

    Included
    50,000
    Overage
    $0.00325/MAU

    SSO MAU

    Included
    —
    Overage
    not available

    Phone MFA + SMS

    MFA
    not available
    Per-SMS
    —
  • Workshop

    Regular MAU

    Included
    100,000
    Overage
    $0.00325/MAU

    SSO MAU

    Included
    50
    Overage
    $0.015/MAU

    Phone MFA + SMS

    MFA
    $70/mo
    Per-SMS
    carrier list × 1.10
  • Studio

    Regular MAU

    Included
    100,000
    Overage
    $0.00325/MAU

    SSO MAU

    Included
    50
    Overage
    $0.015/MAU

    Phone MFA + SMS

    MFA
    $70/mo
    Per-SMS
    carrier list × 1.10
  • Foundry

    Regular MAU

    Included
    100,000
    Overage
    $0.00325/MAU

    SSO MAU

    Included
    50
    Overage
    $0.015/MAU

    Phone MFA + SMS

    MFA
    $70/mo
    Per-SMS
    carrier list × 1.10

Third-Party MAU (external identity providers): $0.000325/MAU on all paid tiers. Not available on Free.

See example bills

Sample monthly bills. Your own usage may vary.

Workshop · Self Serve · 80K MAU, within bundle

  • Compute (Workshop @ Self Serve)

    1 × 730 hr · $0.0799/hr$58.31
  • Storage / egress / requests

    within bundle$0.00
  • Regular MAU

    80,000 (within 100k)$0.00
Total$58.31/mo

Workshop · Self Serve · 130K MAU, 25 SSO, Phone MFA + 5K SMS

  • Compute (Workshop @ Self Serve)

    1 × 730 hr · $0.0799/hr$58.31
  • Storage / egress / requests

    within bundle$0.00
  • Regular MAU overage

    30,000 over 100k · $0.00325/MAU$97.50
  • SSO MAU

    25 (within 50)$0.00
  • Phone MFA enablement

    1 · $70/mo$70.00
  • SMS pass-through (US)

    5,000 messages · $0.0071/msg$35.50
Total$261.31/mo

Foundry · Scale · 250K MAU, 100 SSO, Phone MFA + 30K SMS

  • Compute (Foundry @ Scale)

    1 × 730 hr · $0.3168/hr$231.26
  • Storage / egress / requests

    within bundle$0.00
  • Regular MAU overage

    150,000 over 100k · $0.00325/MAU$487.50
  • SSO MAU overage

    50 over 50 · $0.015/MAU$0.75
  • Phone MFA enablement

    1 · $70/mo$70.00
  • SMS pass-through (US)

    30,000 messages · $0.0071/msg$213.00
Total$1,002.51/mo

PRICING BY ACTION

Per-call costs.

Per-call cost to invoke Powabase's built-in features. LLM inference is billed separately, by token usage with each supported provider. Bring your own LLM API key or pay through Powabase.

Web (paid APIs)

  • web_searchper call
    Free: $0.040Self Serve: $0.030Scale: $0.020

    Standard web search via Exa with relevance-ranked results. It's the default tier and handles most agent lookups.

  • web_search_deepper call
    Free: $0.075Self Serve: $0.05625Scale: $0.0375

    Exa agentic deep search. It runs an iterative multi-step crawl for harder research questions a single lookup can't answer.

  • web_search_deep_reasoningper call
    Free: $0.100Self Serve: $0.075Scale: $0.050

    Exa deep search plus a reasoning pass that synthesizes across sources. Use this most thorough tier for complex research.

  • web_scrapeper page
    Free: $0.010Self Serve: $0.0075Scale: $0.005

    Firecrawl-backed fetch that returns clean markdown for a single page.

Indexing

  • indexing_graphindexper 1k tokens
    Free: $0.010Self Serve: $0.0075Scale: $0.005

    GraphIndex builds an entity-relation graph over the corpus to power multi-hop queries.

  • indexing_doc2jsonper 1k tokens
    Free: $0.008Self Serve: $0.006Scale: $0.004

    Doc2JSON sliding-window extraction that turns long documents into structured records aligned to your schema.

  • indexing_pageindexper 1k tokens
    Free: $0.006Self Serve: $0.0045Scale: $0.003

    PageIndex builds a hierarchical tree over the document, which works best for reasoning over long structured docs.

  • metadata_enrichmentper 1k tokens
    Free: $0.004Self Serve: $0.003Scale: $0.002

    Adds titles, summaries, and tags at indexing time so retrieval can filter and rerank on richer metadata.

  • indexing_chunkembedper 1k tokens
    Free: $0.002Self Serve: $0.0015Scale: $0.001

    ChunkEmbed does classic chunking plus embedding into pgvector. The default and cheapest strategy.

Extraction

  • ocr_pagesper page
    Free: $0.006Self Serve: $0.0045Scale: $0.003

    VLM-based page extraction (LightOnOCR-2, Mistral, or PaddleOCR). We run it by default on every ingested document, so a single file with mixed scanned and text-extractable pages is handled correctly. If you know the source is all native text, opt into faster text-only parsers per ingestion run: OpenDataLoader, PyMuPDF (fitz), or pdfplumber.

Retrieval

  • tree_searchper call
    Free: $0.010Self Serve: $0.0075Scale: $0.005

    PageIndex tree-walk retrieval, which navigates the document hierarchy instead of flat top-k chunks.

  • reranker_callper call
    Free: $0.010Self Serve: $0.0075Scale: $0.005

    Cross-encoder rerank via Cohere, Voyage, Jina, or ZeroEntropy, applied after retrieval to sharpen the top results.

  • hybrid_searchper call
    Free: $0.004Self Serve: $0.003Scale: $0.002

    Vector + BM25 with reciprocal-rank fusion. This is our default retrieval mode.

  • query_enrichmentper call
    Free: $0.004Self Serve: $0.003Scale: $0.002

    Query rewriting or HyDE, which expands the user's question before search to improve recall.

  • bm25_searchper call
    Free: $0.002Self Serve: $0.0015Scale: $0.001

    BM25 / full-text search over your tables. Keyword-precise, and it needs no embeddings.

  • vector_searchper call
    Free: $0.002Self Serve: $0.0015Scale: $0.001

    Plain cosine search over pgvector embeddings. It's the fastest path when you already know the question shape.

Agents & orchestrations

  • orchestration_runper run
    Free: $0.004Self Serve: $0.003Scale: $0.002

    Multi-agent orchestration base fee, covering the supervisor loop that routes between agents.

  • agent_runper run
    Free: $0.002Self Serve: $0.0015Scale: $0.001

    Base fee for a single ReAct agent invocation, before per-tool charges.

  • agent_tool_callper call
    Free: $0.002Self Serve: $0.0015Scale: $0.001

    One tool invocation from an agent, whether a custom HTTP tool, an MCP server, or a generic builtin.

  • agent_tool_code_executeper call
    Free: $0.002Self Serve: $0.0015Scale: $0.001

    Sandboxed code execution called from an agent. The runtime is isolated and there's no infra to manage.

Workflows

  • workflow_runper run
    Free: $0.002Self Serve: $0.0015Scale: $0.001

    Base fee per run of the visual / API workflow engine built into Powabase.

  • workflow_block_codeper call
    Free: $0.002Self Serve: $0.0015Scale: $0.001

    A code block inside a workflow that runs your snippet in the sandboxed code executor.

  • workflow_block_external_apiper call
    Free: $0.002Self Serve: $0.0015Scale: $0.001

    External HTTP block in a workflow, for outbound calls to third-party APIs.