What did Hlido score Dograh?

Dograh scored 76/100 (STEADY) in Hlido's independent, hands-on review.

How much does Dograh cost?

Hlido recorded the pricing model as: Open source. Derived from evidence Hlido holds, not vendor-supplied; verify current prices on the vendor's pricing page.

Does any vendor pay Hlido for placement?

No. Hlido takes no money from the agents it rates — scoring weights stay private and the evidence behind every verdict is public.

Dograh

Name: Dograh review
Item: Dograh
Rating: 76
Author: Hlido Editor

Voice · tested 2026-09-05 · re-test due 2026-12-04 · by the Hlido desk, not the vendor

In short: Open-source, self-hostable voice-agent platform positioned squarely as the Vapi/Retell alternative — unusually broad provider matrix and a genuine air-gapped story, but the voice quality and privacy claims are untested on the public surface.

4 PASS · 1 FAIL of 5 public-surface claims

Visit Dograh → scorecard.json

Quick answer

Dograh scores 76/100 (STEADY) on Hlido’s independent, hands-on test (reviewed 2026-09-05). STEADY (76): the product is clearly articulated, the feature surface is deep and specific (not vaporware bullet points), the open-source license and traction signals are verifiable, and the positioning fills a real gap ( Pricing: Open source (free entry point documented).

Hlido Editor · 2026-09-05

Dograh makes the sharpest positioning bet in the voice-agent space right now: it is the fork-it-yourself, run-it-on-your-own-infrastructure answer to the hosted incumbents (Vapi, Retell). The public surface backs the pitch with substance rather than slogans — a visual workflow builder where every stage of the pipeline (inbound channel, STT, LLM, TTS, telephony) is a swappable node, and a genuinely long list of interchangeable providers (Deepgram, ElevenLabs, Cartesia, Groq, OpenRouter, Twilio, Telnyx, Plivo, Asterisk, and dozens more), plus a speech-to-speech path that skips the cascade entirely. The BSD-2 license and 5.5k+ GitHub stars make the open-source claim credible, and the Product Hunt #1-of-day/week recognitions and NVIDIA Inception listing are real external signals. The load-bearing differentiator — 'your data never leaves', deploy on-prem or air-gapped so no calls, recordings, transcripts or model inference cross your boundary — is exactly what a regulated buyer wants, and exactly what cannot be confirmed from a marketing page: it is a claim to verify in a deployment, not a fact we observed. For an agentic ecosystem, the appeal is that a coding agent (the site names Claude Code, Cursor, and others) can reportedly spin up and modify agents against your own stack. Treat that, too, as an advertised capability pending hands-on confirmation. The honest read: a serious, well-differentiated OSS project with strong momentum, reviewed here only at the surface.

Why STEADY

STEADY (76): the product is clearly articulated, the feature surface is deep and specific (not vaporware bullet points), the open-source license and traction signals are verifiable, and the positioning fills a real gap (self-hosted/air-gapped voice agents). Held below the top band because the defining claims — voice quality, latency, and the air-gapped/no-data-egress guarantee — are unverified at Tier-1, pricing for the hosted 'Try Cloud' tier was not captured, and a young high-velocity project carries more roadmap risk than the established hosted incumbents.

Public-surface checklist

PASS Homepage loads (required)
PASS Primary value prop (required) — 'The self-hostable voice agent platform · OSS alternative to Vapi & Retell'
PASS Cta present (required) — 'Try Cloud'
FAIL Pricing or access — Pricing link present but tier detail not captured at Tier-1
PASS Evidence or demo — Interactive pipeline/'Start call' demo shown on homepage

What it does well

Sharp, credible positioning as the self-hostable OSS alternative to Vapi/Retell
Genuinely broad, swappable provider matrix across STT, LLM, TTS and telephony
Visual workflow builder plus a speech-to-speech path for lower-latency pipelines
Open-source (BSD-2) with strong external traction signals (5.5k+ stars, Product Hunt #1)
On-prem / air-gapped deployment story that hosted incumbents structurally cannot match

What it fails at

Core voice quality, latency and reliability are untested on the public surface
The headline 'data never leaves / air-gapped' guarantee is advertised, not verified here
Hosted 'Try Cloud' pricing and limits were not captured at Tier-1
Young, fast-moving project — API/feature stability and long-term maintenance are unproven
Agent-runtime integration (Claude Code/Cursor spinning up agents) is a claim, not a tested path

Best for

Teams that need voice agents inside their own VPC or fully air-gapped for compliance
Engineering orgs that want to own and fork the stack rather than rent a hosted API
Builders who want to swap STT/LLM/TTS/telephony providers without re-platforming
Anyone evaluating an OSS exit from Vapi/Retell lock-in

Not recommended for

Teams that want a fully managed, zero-ops hosted voice API and nothing to run
Buyers who need a vendor with a long production track record today
Use cases that require an independently benchmarked voice-quality/latency guarantee before commit

Pricing & access

ModelOpen source
Free entry pointYes — a free tier or open-source edition is documented
Pricing findable on the public surfaceFAIL Pricing link present but tier detail not captured at Tier-1 (tested 2026-09-05)

Derived from Hlido-held evidence only (engine checklist + editorial text); quotes are verbatim from the scorecard; not vendor-supplied; re-derived daily. Verify current prices on the vendor's pricing page. Last verified 2026-09-05.

Compared to

Vapi self-hosting-and-data-residency
Vapi is the polished hosted incumbent with a longer track record; Dograh is the self-hostable OSS challenger. Choose Vapi for zero-ops managed infrastructure; choose Dograh when data residency, forking, or air-gapped deployment is the constraint.
Retell AI open-source-vs-hosted
Retell is another strong hosted voice-agent platform. Dograh's pitch is explicitly the open-source alternative to it — same problem space, opposite deployment philosophy.
Bland AI provider-flexibility
Bland is hosted and phone-call-focused. Dograh trades turnkey convenience for provider flexibility and on-prem control.

Agent relevance

API Webhook Behavioral-testable

Self-hosted voice-agent platform with a workflow builder and provider-pluggable pipeline; the vendor claims agent runtimes (Claude Code, Cursor) can build and deploy agents against it. Behaviourally testable in a deployment, though not exercised at Tier-1.

Agent-friendly score: 6/10

Evidence

Positions as open-source, self-hostable alternative to Vapi & Retell — source (2026-09-05) verified
Swappable pipeline (STT/LLM/TTS/telephony) with a broad provider matrix — source (2026-09-05) verified
Open source under BSD-2-Clause — source (2026-09-05) verified
Data never leaves your boundary / on-prem / air-gapped inference — source (2026-09-05)
Point Claude Code/Cursor/agent runtimes at it to build & deploy agents — source (2026-09-05)

scorecard.json · registry · methodology

More: compare agents · best of · developer tools · incident registry

Verdict by Hlido Editor, our automated editorial system · Method: public-surface-tier-1+editorial-narrative-v2 · Methodology version 2026.05 · Next review due 2026-12-04

How this page was produced. The scores, claim verdicts and evidence come from automated hands-on testing of the product’s public surface. The written analysis is drafted by an AI system, and pages publish without a person reviewing each one. Hlido publishes this record and answers for it — tell us if anything here is wrong and we will correct it.

Embed this trust badge

Live, always-current independent score — free to embed on your site or README. No vendor pays for placement.

Markdown

[![Hlido trust score](https://hlido.eu/badge/dograh.svg)](https://hlido.eu/check/?agent=dograh)

HTML

<a href="https://hlido.eu/check/?agent=dograh"><img src="https://hlido.eu/badge/dograh.svg" alt="Hlido trust score"></a>