# Cekura — Voice-agent testing that connects simulations, production monitoring, and improvement loops.

> Source: The Agents Index — https://theagentsindex.com/cekura (structured, researched, re-verified)

Cekura is a voice-agent testing and observability platform for simulated conversations and production-call monitoring. It evaluates conversational quality, tool calls, and latency to help teams improve their agents.

**Best for:** Voice AI teams testing, monitoring, and improving conversational agents before production

| Fact | Value |
| --- | --- |
| Website | https://cekura.ai/ |
| Category | Voice agent testing |
| Open source | No |
| Price | Free |
| Price source | https://www.cekura.ai/pricing |
| Price checked | 29 October 2026 |
| MCP server | Yes |
| API | Not stated |
| Self-host | Partly, on Enterprise |
| Facts | 17 facts from 6 sources |
| Facts checked | 29 October 2026 |

Screenshot of Cekura: https://theagentsindex.com/img/1920/screenshots/cekura.webp, captured 1 November 2026

**What we checked**

- Pages last read: 29 October 2026
- Facts checked: 17 facts (From 6 sources)
- AI answers: Named in 0 of 151 (Our October check)

## What we checked on Cekura

### Which models it runs on

**Model choice** · Yes — Teams can run identical scenarios across platforms and models, then select the option that performs best.

> Run the same scenarios across platforms and models. Pick the one that actually performs.
> — https://www.cekura.ai/ · Homepage, Benchmarking section, checked 29 October 2026

### Whose machine it runs on

**Self-hosted** · Partly — VPC hosting and on-premises deployment are documented for Enterprise, rather than the self-serve plans.

> VPC hosting / on-prem | — | — | Both
> — https://www.cekura.ai/pricing · Compare Plans table, VPC hosting / on-prem row, checked 29 October 2026

### Building with it

**SDK languages** · Yes — The LiveKit tracing integration documents Cekura SDK installation and usage in Python and JavaScript.

> Install the Cekura SDK **Python**
> — https://docs.cekura.ai/documentation/integrations/livekit/tracing · LiveKit Tracing documentation, Setup step 1, checked 29 October 2026

**Tracing and evals** · Yes — LiveKit tracing captures transcripts, tool calls, performance metrics, session logs, and OpenTelemetry traces for analysis.

> Complete conversation transcripts with full message history
> — https://docs.cekura.ai/documentation/integrations/livekit/tracing · LiveKit Tracing documentation, Overview / What you get, checked 29 October 2026

### What it costs to run

**How it meters** · Yes — The entry pay-as-you-go plan meters voice testing by minute and production monitoring by individual call.

> $0.25 per minute $0.025 per reply $0.05 per monitored call
> — https://www.cekura.ai/pricing · Compare Plans table, Usage rates row, Pay as you go column, checked 29 October 2026

**Cost of one job** · Yes — On the pay-as-you-go tier, a monitored call costs $0.05 and voice testing costs $0.25 per minute.

> $0.25 per voice testing minute
> — https://www.cekura.ai/pricing · Pay as you go plan card, What's included, checked 29 October 2026

**Free tier** · Partly — You can sign up with no card and get 300 credits, about 60 minutes of voice testing, but there is no standing free allowance after that: every further minute, reply and monitored call is billed.

> 300 free credits to start ≈ 60 min
> — https://www.cekura.ai/pricing · "Every plan includes" strip, below the three plan cards, checked 28 October 2026

**MCP server** · Yes — A hosted MCP server at api.cekura.ai/mcp authenticates by OAuth and gives an assistant tools to list agents, create metrics, generate evaluators and run scenarios, with regional endpoints for the US, India and the EU.

> MCP gives your assistant permissioned tools to list agents, create metrics, generate evaluators, run tests, inspect transcripts, and analyze results.
> — https://docs.cekura.ai/mcp/overview · "Why install both?" section, "MCP adds action" card, checked 28 October 2026

**Bring your own key** · Yes — Client-side redaction can use the customer's LLM and speech-to-text providers with the customer's API keys.

> with your own AI providers, before anything reaches Cekura
> — https://docs.cekura.ai/documentation/guides/observability/client-side-redaction · Client-Side Redaction documentation, page introduction, checked 29 October 2026

### Buying it for a team

**A company can buy it** · Yes — Cekura sells a Startup plan for voice AI teams and an Enterprise plan for complex deployments.

> Predictable monthly capacity for voice AI teams ready to scale.
> — https://www.cekura.ai/pricing · Startup Plan card, description, checked 29 October 2026

**Seat model** · Yes — The pay-as-you-go plan includes one free seat and charges $30 monthly for each additional seat; Startup includes ten.

> 1 free · $30/mo each after
> — https://www.cekura.ai/pricing · Compare Plans table, Seats row, Pay as you go column, checked 29 October 2026

**Admin controls** · Partly — Enterprise provides multiple projects with access control, while the public plan table does not offer that control on lower plans.

> Multiple projects with access control
> — https://www.cekura.ai/pricing · Enterprise plan card, What's included, checked 29 October 2026

**Audit log** · Partly — Audit logs are listed with the Enterprise deployment and governance capabilities, not the self-serve plans.

> VPC / on-prem · SSO · SCIM · audit logs
> — https://www.cekura.ai/pricing · Enterprise plan card, What's included, checked 29 October 2026

**Single sign-on** · Partly — SAML SSO and SCIM are available on Enterprise, while the two lower plans show no SSO offering.

> SAML SSO + SCIM
> — https://www.cekura.ai/pricing · Compare Plans table, SSO / SCIM row, Enterprise column, checked 29 October 2026

### What happens to your code

**Data residency** · Partly — Enterprise includes data residency controls, but the public plan table does not identify the available jurisdictions.

> Audit logs · IP allowlist · data residency
> — https://www.cekura.ai/pricing · Compare Plans table, Audit logs · IP allowlist · data residency row, Enterprise column, checked 29 October 2026

**Getting out** · Yes — The privacy notice gives users a route to request access to a copy of their personal information or request deletion.

> To request to review, update, or delete the data we collect from you
> — https://www.cekura.ai/privacy-policy · Privacy Policy, section 15, checked 29 October 2026

**Certifications** · Yes — Cekura labels every plan HIPAA and GDPR, while Startup includes a signed BAA and DPA.

> Signed BAA & DPA
> — https://www.cekura.ai/pricing · Startup Plan card, What's included, checked 29 October 2026

## Cekura price and terms

**Terms**

- Free plan: Yes — One seat is free — 300 credits are supplied to start with no credit card, while continued usage is metered.

**What needs a higher plan**

- Enterprise: Self-hosted, Admin controls, Audit log, Single sign-on, Data residency

## For AI agents

- MCP server: Yes

Pricing page as we captured it: https://theagentsindex.com/img/1920/screenshots/cekura/pricing.webp, captured 1 November 2026

Read on https://www.cekura.ai/pricing, 29 October 2026.

## Cekura alternatives

- [Hamming AI](https://theagentsindex.com/hamming-ai) · Price on request: Voice and chat agent QA combining simulated testing, red-teaming, and production-call monitoring.

It publishes no price.

[1 alternative](https://theagentsindex.com/cekura/alternatives/)

## Questions about Cekura

### How much does Cekura cost?

Cekura is free to use. From cekura.ai/pricing, checked 29 October 2026.

### Does Cekura have an API or an MCP server?

It has an MCP server.

### Is Cekura open source, and can you host it yourself?

You can host it yourself on Enterprise.

### What have we checked on Cekura?

17 facts, under these headings: Which models it runs on, whose machine it runs on, building with it, what it costs to run, buying it for a team, what happens to your code.

### How do we check Cekura?

We hold 17 facts on Cekura, quoted from 6 pages. Facts and price last checked 29 October 2026.

How we check every listing: https://theagentsindex.com/methodology/

## Sources and changes

Last changed 1 November 2026

**Sources** (6)

- https://www.cekura.ai/pricing, read 29 October 2026
- https://docs.cekura.ai/documentation/integrations/livekit/tracing, read 29 October 2026
- https://www.cekura.ai/, read 29 October 2026
- https://docs.cekura.ai/documentation/guides/observability/client-side-redaction, read 29 October 2026
- https://docs.cekura.ai/mcp/overview, read 28 October 2026
- https://www.cekura.ai/privacy-policy, read 29 October 2026

## Is this yours?

AI assistants like ChatGPT and Claude opened The Agents Index pages 2,746 times in September. Claim Cekura: https://theagentsindex.com/claim/cekura

Report an error if a fact is wrong: hello@theagentsindex.com, subject "Report an error: Cekura (cekura)".
