Choosing Dograh vs Bolna comes down to how much of the platform you own. Bolna rents an India-first hosted platform on an MIT orchestration core, with its preferred models bundled into one per-minute rate. Dograh, an open-source AI voice agent platform, includes its visual builder in the open code, and you can self-host it or run it on Dograh Cloud.
Key Takeaways
- Both publish open-source code, but only Dograh's includes the visual builder.
- Bolna's 6¢ minute includes its preferred models, while Dograh prices models separately.
- Dograh's Transfer Call does not work on Exotel or Vobiz numbers.
This post is part of our guide to Bolna alternatives, where Dograh is the alternative whose open-source code includes the builder. Here we compare Dograh and Bolna row by row and explain what each price covers. Then we say plainly which teams should stay on Bolna.
Dograh and Bolna, row by row
Bolna rents a hosted platform on an open engine, while Dograh is the whole platform as open-source software you can run anywhere.
| Row | Bolna | Dograh |
|---|---|---|
| License | MIT orchestration core; hosted APIs and no-code dashboard closed source | BSD-2-Clause, the whole platform, builder included |
| Where it runs | Bolna's cloud, US by default; India residency and on-prem containers on Enterprise | Your own servers in any country, Dograh Cloud, or on-prem and private cloud on Enterprise |
| How you pay | Prepaid wallet in rupees or dollars; a per-minute rate with preferred models included, telephony and platform fee added | Free to self-host; Dograh Cloud bills a per-minute platform fee, with models and phone billed separately |
| Calls at once | From 10 outbound on paid accounts, scaling with usage; inbound not capped; up to 25 on a Pilot | Dograh Cloud includes 10, more on request; self-hosted runs 10 by default, a limit you can raise |
| Models | Bundled preferred models, your own keys, or your own OpenAI-compatible LLM | Dograh's models, any supported provider with your own keys, or open-weight models on your own servers |
| Builder | No-code dashboard with conversation, graph and IVR agents, plus Studio | Visual workflow builder in the open code, plus an MCP server |
| Phone carriers | Twilio, Plivo, Exotel, Vobiz, Airtel, Acefone, Telnyx, VoiceLink or your own SIP trunk | Twilio, Plivo, Telnyx, Vonage, Cloudonix, Exotel, Vobiz or Asterisk ARI |
| Transfer to a person | A Transfer Call tool in the dashboard | Transfer Call on Twilio, Telnyx, Plivo, Cloudonix and Asterisk ARI, not on Exotel, Vobiz or Vonage |
| Keypad input | Yes, through IVR agents | No keypad (DTMF) tool; callers answer by voice |
| Languages | More than 10 Indian languages and 20+ in all | 70+ languages, depending on the speech-to-text and voice providers you choose |
| Security | An A+ rating on its latest VAPT security test | Code your security team can read; SSO, SAML and audit logs on Enterprise |
| Support | A Pilot month with a WhatsApp channel and four 1:1 calls; an account manager on Enterprise | A dedicated account manager and custom SLAs on Enterprise |
Open code, and where each one runs
Both companies publish open-source code, and the difference is how much of the product it covers. Bolna's repository holds its orchestration core under MIT, a Python framework where agents are written as JSON, with no UI. Its hosted APIs and no-code dashboard are closed. Dograh's repository holds the whole platform, the same one that runs as Dograh Cloud. Our post on the open-source alternative to Bolna compares the two repositories in detail.
Hosting follows from that. Hosted Bolna runs in the US by default, and India data residency is an Enterprise feature. Bolna's on-prem containers also come with Enterprise, and they send usage metrics to Bolna's cloud for billing.
Dograh runs wherever you put it. You can self-host it with Docker on servers in any country, with no sales contract. For Indian teams, data stays in India by running Dograh on servers in India, either yourself or through Dograh's managed on-prem plan.
Models and the builder
Both platforms let you pick models, and Dograh lets you run them on your own servers too. Dograh is the orchestration layer that runs the call flow, tools and telephony, and the speech and language models are a separate layer you choose. That can be Dograh's own models, any supported provider with your own keys, or open-weight models through Ollama, vLLM or Speaches.
Sarvam and Smallest AI are providers in both, which matters for Indian-language calls. Bolna's list also covers Soniox, Anthropic and Gemini, and its speech-to-speech agents run on OpenAI Realtime or Gemini Live. Dograh runs speech-to-speech on OpenAI Realtime, Gemini Live, Azure, Grok or Ultravox. Hindi is one of Dograh's Language Hints, used with Dograh's models or Deepgram Flux. Our post on multilingual voice agents covers languages across providers.
Both have a visual builder now. Bolna's graph agents got a visual editor in its dashboard in June 2026, and Studio drafts an agent from a brief. Dograh's visual workflow builder joins steps with pathways, and a language model can draft the starting workflow from your use case. The difference is where the builder lives. Dograh's ships in the open-source code, so a self-hosted team keeps it, and an AI assistant can create and edit agents through Dograh's MCP server.
Open Source Alternative to Vapi / Retell
Self-hosted voice agent platform — no per-minute fees
dograh-hq/dograh
Star on GitHub
What each price covers
Bolna sells a minute with the models already inside it, so its number and Dograh's measure different things. Bolna's pricing page lists a standard rate of 6¢ (₹5.52) a minute. With its preferred models, that covers speech to text, the language model and the voice. Telephony and Bolna's platform fee are added per minute, and larger top-ups bring the rate to about 4.5¢. Pilot plans are $300 for 6,500 minutes or $500 for 12,000.
Dograh keeps the platform and the models on separate lines. Self-hosted Dograh has no per-minute platform fee (servers, model keys and phone minutes still cost). Dograh Cloud has a 1¢-a-minute platform fee, and Dograh's own models are about $0.07 a minute for most models, as listed on Dograh's pricing page. With your own keys, each model provider bills you directly, and your carrier bills the phone line.
Bolna supports your own keys too, and then you pay the providers plus Bolna's platform fee. So compare the whole minute with the same models on both sides, and add the servers and engineering time if you self-host. Our guide to bring your own keys or managed models explains that choice.
Who should stay on Bolna
Bolna is the better fit for some teams, and the reasons are concrete. Stay if you want one prepaid per-minute price with the models included, billed in rupees. Stay if you need more than 10 Indian languages ready to use, or want the vendor itself to host your data in India on its Enterprise plan.
Stay too if your calls run on Exotel or Vobiz and must reach a person. Dograh's Transfer Call works on Twilio, Telnyx, Plivo, Cloudonix and Asterisk ARI, and not on those two carriers. Bolna's IVR agents take keypad input, which Dograh does not. And a Pilot with a month of WhatsApp support suits a team that wants the vendor close at hand while it launches.
When Dograh is the stronger fit
Dograh is the stronger fit for teams that want to own the platform and choose every model. Its code is open, so your security team can read what it approves, and you can change it or keep running your version whatever happens upstream. It runs in any country you choose, and every model is your choice, from Sarvam's voices to local open-weight models.
When it transfers, Dograh calls the destination first and connects the caller only when someone answers. On Twilio, it can play a short spoken introduction to both people before connecting them. No summary goes across by default, so the agent writes the caller's details to your system first. QA runs on test and live calls and scores each step of the call separately.
Self-hosting puts the servers and their upkeep on your team, and telemetry is on by default, so run ENABLE_TELEMETRY=false ./start_docker.sh and delete POSTHOG_API_KEY from docker-compose.yaml to switch it off. There is no built-in SMS, and CRMs connect through HTTP API tools and webhooks. Run your real call script on both over a phone line, since most phone calls still reach the agent as 8 kHz audio while a browser demo runs at 16 kHz. To try Dograh, self-host it from GitHub, or start on Dograh Cloud.
Join the Dograh Community
Dograh is an OSS alternative to Vapi. Join our Slack community for queries, releases, best practices & community interactions.
Glossary
- Preferred models
- The speech-to-text, language and voice models included in Bolna's standard per-minute rate. Using your own keys instead moves those costs to your own provider accounts, plus Bolna's platform fee.
- Platform fee
- A per-minute charge for running the call on a platform, separate from the models and the phone line. Bolna adds one to its rate, and Dograh Cloud charges one while self-hosted Dograh does not.
- Graph agent
- A Bolna agent built as connected steps rather than one prompt, edited visually in Bolna's hosted dashboard. Dograh's equivalent is a workflow of steps joined by pathways in its open-source builder.
