# OmniDimension — No-code platform for building voice AI agents and running them on phone, web chat, WhatsApp, email and SMS.

> Source: The Agents Index — https://theagentsindex.com/omnidimension (structured, researched, re-verified)
> Facts last verified: 2026-09-09

OmniDimension is a no-code platform for building voice AI agents from a text prompt and deploying them across phone, web chat, WhatsApp, email and SMS. It bundles telephony, a visual workflow builder, a CRM, call analytics and a REST API.

| Fact | Value |
| --- | --- |
| Website | https://omnidim.io |
| Price | from $15/month billed monthly, $12.50/month billed annually |
| Free trial | Partly |

Complimentary sign-up credits of an unpublished size, no card required; there is no standing $0 plan. The cheapest standing routes are a pay-as-you-go top-up from $14 or Starter at $15/month.

## What it does

**Done-for-you service** · Partly — You run the software yourself from a no-code dashboard, but the terms of use offer paid professional services on request, and the Enterprise tier adds a dedicated account manager, priority onboarding and live training.

> OmniDimension offers customized professional services which include personalized consultations, advanced implementation assistance, optimization strategies, and more, tailored to Your specific needs.
> — https://omnidim.io/terms-of-use · Section 1.2 Professional Services

## How much it does unattended

**Runs autonomously** · Yes — An agent runs whole inbound or outbound calls unattended, and workflows fire calls, messages and CRM writes from a trigger with nobody watching.

> You connect a trigger (something that starts the flow) to a sequence of actions (calls, messages, CRM updates, and logic), so work happens automatically without writing code.
> — https://docs.omnidim.io/docs/workflows · OmniWorkflows, opening description

**Multi-agent** · Partly — Workflows chain agent actions and can call sub-workflows, so several agents run in sequence, but the docs describe no agent-to-agent handoff or shared-state orchestration.

> Flow control: Condition, Switch, Delay, Wait for time, Time condition, Run workflow (sub-workflow), Code, Note
> — https://docs.omnidim.io/docs/workflows · What a workflow is made of, Actions list

**Agent permissions** · Partly — The agent acts alone inside limits you write in advance: natural-language transfer and hang-up conditions, a hard call-length ceiling and per-campaign calling hours, so escalation is pre-configured rather than requested mid-call.

> Conditional call transfer to a human agent or another number.
> — https://docs.omnidim.io/docs/api-reference/agents/createAgent · Agent configuration schema, transfer object description

## Which models it runs on

**Claude** · No — The pricing page publishes the full model catalogue and its alphabetical LLM list runs from Bedrock straight to Gemini with no Anthropic or Claude entry; the agent API model enum likewise names none.

> Bedrock:mistral.mistral-large-3-675b-instruct Gemini-2.5-flash Gemini-2.5-flash-lite Gemini-3.5-flash-lite Gemini-3.6-flash Gpt-3.5-turbo Gpt-4.1-mini
> — https://omnidim.io/pricing · Flexible model selection panel, Show available models, LLM list (alphabetical, where Claude would fall between Bedrock and Gemini)

**GPT** · Yes — OpenAI models are the largest block in the published catalogue, offered both directly and through Azure, and the agent API accepts them by name.

> Gpt-4o Gpt-4o-mini Gpt-5-mini Gpt-5-nano Gpt-5.1 Gpt-5.4-mini Gpt-5.4-nano Gpt-5.6-luna
> — https://omnidim.io/pricing · Flexible model selection panel, Show available models, LLM list

**Open models** · Yes — Open-weight families are selectable through the OpenRouter and Bedrock entries in the catalogue, so you are not confined to closed frontier models.

> Openrouter:google/gemma-4-31b-it Openrouter:minimax/minimax-m2.7 Openrouter:nvidia/nemotron-3.5-lightning Openrouter:openai/gpt-oss-120b
> — https://omnidim.io/pricing · Flexible model selection panel, Show available models, LLM list

**Model choice** · Yes — You choose the speech-to-text, LLM and text-to-speech provider per agent and may mix them freely; the vendor does not pin you to one stack.

> You can use any combination of supported models for your Voice AI agents.
> — https://omnidim.io/pricing · Flexible model selection section, above the Show available models control

## Where you use it

**In your editor** · Partly — There is no editor extension, but the hosted MCP server plugs the agent tools into Cursor, VS Code and Windsurf, so an IDE drives the platform rather than hosting it.

> Cursor, VS Code, Windsurf | Hosted. These clients speak HTTP MCP natively.
> — https://docs.omnidim.io/docs/mcp · Which one should I use? table, Cursor/VS Code/Windsurf row

**On the command line** · Partly — The only terminal-run component is the local MCP server, installed and diagnosed with npx sub-commands; no CLI exists for building agents or dialling calls on its own.

> It runs on your machine as a stdio child process, reads your OmniDimension API key from an environment variable, and proxies tool calls to the OmniDimension REST API.
> — https://docs.omnidim.io/docs/mcp/local-server · Local MCP server, opening paragraph

**In your pipeline** · Partly — The vendor points CI and headless automation at the local API-key MCP server and the REST API, but nothing in the product acts on a pull request.

> Use the local server when you want: An API-key flow instead of OAuth (CI, automation, scripts).
> — https://docs.omnidim.io/docs/mcp/local-server · Local MCP server, 'Use the local server when you want' list

**In a browser** · Yes — Everything is built in a hosted dashboard, and the same agent can be embedded on your own site as a chat and voice widget with a no-code script.

> Web chat and voice widget: embed the agent on your website.
> — https://docs.omnidim.io/docs/omnichannel · Channels you can deploy to list

## Whose machine it runs on

**Self-hosted** · No — No self-hosted, VPC or on-premise option appears anywhere in the pricing tiers or docs; even the white-label agency product runs on the vendor's infrastructure with the customer only pointing DNS at them.

> Once your domain points at us, we issue and renew the HTTPS certificate for you automatically. There is nothing to buy or install.
> — https://docs.omnidim.io/docs/omnirelay/faq · Setup and domains, HTTPS certificate answer

**Open source** · Partly — The two MCP servers and the Python, TypeScript and web SDKs are public and permissively licensed, while the platform itself is proprietary and the terms forbid reverse engineering it.

> omnidim-mcp-cloud (Apache 2.0). Hosted OAuth + HTTP server (FastAPI, Python).
> — https://docs.omnidim.io/docs/mcp · Open source section, GitHub repository list

## What it costs to run

**How it meters** · Yes — Billing is a flat monthly prepaid balance drawn down per minute of conversation, with per-message rates for chat and WhatsApp and published overage rates on every tier.

> Each tier lowers your per-minute voice rate and unlocks the next set of capabilities. Pay a flat monthly price only for the minutes you talk.
> — https://omnidim.io/pricing · Voice AI pricing section, introduction above the plan cards

**Free tier** · Partly — You can sign up and build without a card, but the free entry is a one-off grant of complimentary credits rather than a standing zero-dollar plan; the cheapest standing route is a pay-as-you-go top-up.

> Yes, OmniDimension offers complimentary credits for new users to explore our platform. You can upgrade to a paid subscription when you're ready to scale.
> — https://omnidim.io/pricing · FAQs, 'Is there a free trial?'

**API access** · Yes — A documented REST API with a Bearer key covers agents, calls, campaigns, numbers and analytics, with Python, TypeScript and browser SDKs and a downloadable OpenAPI spec.

> The OmniDimension API lets you create agents, dispatch calls, manage phone numbers, and pull call analytics from your own systems. Auth is a single Bearer API key on every request.
> — https://docs.omnidim.io/docs/api-reference · API reference, opening paragraph

**MCP server** · Yes — Two MCP entry points expose every REST endpoint as a tool: a hosted OAuth server and a local stdio package, so another agent can drive calls and campaigns.

> OmniDimension ships two MCP entry points to the same set of tools (agents, calls, bulk calls, phone numbers, knowledge base, providers)
> — https://docs.omnidim.io/docs/mcp · Overview, opening paragraph

**Bring your own key** · Partly — You bring your own telephony and messaging accounts, importing numbers from Twilio, RingCentral, Vonage or Exotel and WhatsApp via your own Meta credentials, but the agent model config offers only the vendor's own catalogue, with no field for your own LLM key or cloud.

> Bring your existing phone numbers from Twilio, RingCentral, Vonage, or Exotel.
> — https://omnidim.io/ · Ecosystem section, Telephony card

## Buying it for a team

**A company can buy it** · Yes — Plans are sold to companies with named team-member counts, and the Enterprise tier is bought through sales with unlimited seats, a custom DPA and a security review.

> Custom concurrency and unlimited seats
> — https://omnidim.io/pricing · Enterprise plan card, 'Everything in Growth, plus' list

**Seat model** · Yes — Seats are not billed individually; each tier includes a fixed ceiling of team members, one on Starter rising to ten on Growth, and only Enterprise removes the cap.

> Team member 1 2 5 10 Custom
> — https://omnidim.io/pricing · Compare plans table, Plan details, Team member row

**Pooled budget** · Yes — The monthly price is one prepaid balance for the whole account that voice minutes and messages draw from, so team members spend from a shared pot rather than personal allowances.

> Monthly price is prepaid balance for voice and messages. Your subscription controls features; top-ups lower your rate but do not unlock features.
> — https://omnidim.io/pricing · FAQs, 'How do plan credits and top-ups work?'

**Admin controls** · Partly — Real per-user controls exist only for reseller accounts, which can switch dashboard menus on or off per client and cap their concurrency; ordinary team plans have no documented roles or permission settings, and agent limits are set per agent instead.

> Enable or disable dashboard menu access flags for a child user. Only the flags you pass are changed. Flags outside your reseller's permissions are silently ignored.
> — https://docs.omnidim.io/docs/api-reference/reseller/setUserAccessControl · Update user access control, endpoint description

**Audit log** · Partly — Configuration changes are recorded as immutable agent versions with diffs and named automatic backups, but they are purged after about ninety days and there is no account-wide log of administrative or sign-in activity.

> Saved versions are kept for about 90 days and then removed automatically, so your history stays tidy.
> — https://docs.omnidim.io/docs/dashboard-guides/agent-versioning-and-publishing · Save a version section, version-retention callout

**Single sign-on** · Partly — Single sign-on is sold only in the Enterprise bundle alongside the custom DPA and security review; no SAML or OIDC setup is documented, and no self-serve tier includes it.

> SSO, custom DPA and security review
> — https://omnidim.io/pricing · Enterprise plan card, 'Everything in Growth, plus' list

## What happens to your code

**Opt out of training** · Partly — The terms take call recordings and user content for model development unless you opt out of recording, so the control exists but is opt-out, is bundled with recording itself, and is qualified only by de-identification.

> process communication data (“Communications”) and User Content for offering AI‑powered analytics and the development, training, and improvement of artificial intelligence and machine‑learning models
> — https://omnidim.io/terms-of-use · Section 'Call Recording and Third-Party Vendors', first sentence (opens 'If You do not opt-out of recording')

**Data residency** · Partly — The vendor is a US company that processes data in the United States under California law; no EU, UK or India region choice is offered anywhere, despite INR billing.

> If you access the Service from outside the United States, please be aware that your information may be transferred to and processed in the U.S.
> — https://omnidim.io/privacy-policy · Section 8, International Transfers

**Getting out** · Partly — Unused balance survives plan changes and top-ups with no published expiry, and transcripts and recordings can be pulled through the call-logs API, but deletion is a manual email request and no bulk account export is documented.

> Plans lower your per-minute rate and unlock more features, and you can upgrade anytime. Your balance carries over.
> — https://omnidim.io/pricing · FAQs, 'Can I use OmniDimension without a subscription?'

**Certifications** · Partly — No certification is published: security assurances are sold as an Enterprise-only custom DPA and security review, and the vendor's own healthcare page states its HIPAA attestation is not published.

> SSO, custom DPA and security review
> — https://omnidim.io/pricing · Enterprise plan card, 'Everything in Growth, plus' list

## Provenance

Every fact above comes from the vendor. Nothing here is independently corroborated yet.

## Coverage elsewhere

None recorded. Every fact above comes from the vendor and is not independently corroborated yet.

## What changed

- 2026-09-09 — Capabilities · Commercial terms · Free tier · Price

## Correcting this record

Anything marked "Not established" is a question we have not answered — it is not a No. If you represent OmniDimension, or anything here is wrong or out of date, you can supply it with a source: the operations this site accepts from an agent are listed at https://theagentsindex.com/.well-known/agents.json — that catalogue needs no key to read, and each entry states its own auth.

Supplying evidence is free. Nothing that can be bought here changes what a field says, whether a gap is closed, or how this record is scored.