Resumen / Puntos clave
- Bland charges $0.14 a minute (Start) or $0.12 plus $299 a month (Build), covering the LLM, speech and voice but not the phone line.
- Warm and live transfers, SMS, the HIPAA BAA, SSO and self-hosting are Enterprise-only on Bland.
- Telnyx has the lowest stated base rate at $0.05 a minute including speech and voice.
- Retell AI, Vapi, ElevenLabs Agents and Telnyx let you choose your LLM; Bland runs its own.
- LiveKit Agents and Pipecat are open source, so you can self-host without an enterprise contract.
Short answer: Bland charges $0.14 a minute on its free Start plan, or $0.12 on Build with a $299 monthly platform fee, and that rate covers the model, speech recognition and voice but not the phone line. If you want a lower base rate, Telnyx charges $0.05 a minute including speech and voice, with LLM tokens and telephony on top. If you want to pick your own LLM, Retell AI ($0.07 to $0.31 a minute) and Vapi ($0.05 plus providers at cost) both let you. For voices, ElevenLabs Agents comes from the company that makes them. To self-host without an enterprise contract, use open-source LiveKit Agents or Pipecat. Updated November 9, 2026.
How we checked: Bland's plans come from its pricing page and documentation, and every other price from that vendor's own pricing page, all read November 9, 2026. We did not test call quality or latency, and we say nothing about either. For a full-stack cost comparison at 1,000 and 10,000 minutes a month, see our AI voice agent platforms comparison.
Bland AI pricing and plans in November 2026
| Plan | Per minute | Platform fee | Concurrent calls | Daily cap | Voices |
|---|---|---|---|---|---|
| Start | $0.14 (transfers $0.05) | $0, no card | 10 | 100 calls | 1 |
| Build | $0.12 (transfers $0.04) | $299/month | 50 | 2,000 calls | 5 |
| Enterprise | Custom | Contracted to your volume | Custom | Unlimited | Unlimited, plus custom voice actor |
The Start plan comes with "2 credits + an inbound number ($15/mo value)". Extra numbers are $15 a month, per Bland's API docs, or you can bring your own Twilio account or SIP trunk. One inconsistency: Bland's docs summary says the per-minute price covers telephony, while the pricing page says telephony is billed separately at pass-through cost. We go by the pricing page.
Why people compare Bland
- Price per minute. Bland's $0.14 is all-in for the AI, which makes it simple to budget but not the lowest rate. Bland's own FAQ argues that Vapi and Retell stacks reach $0.13 to $0.30 and $0.11 to $0.25 a minute once you add providers. Vapi's calculator puts its own range at $82 to $129 per 1,000 minutes, and Retell's example is $0.11, so check each vendor's calculator with your own model and voice.
- Enterprise focus. Bland's comparison table marks warm and live transfers, SMS and web chat, the BAA for HIPAA, SSO and data residency as Enterprise only. Enterprise deployments follow a 28-day framework with forward-deployed engineers. That suits a contact center; a small team that needs a warm transfer has to talk to sales.
- Voices. Start allows 1 voice and Build 5; unlimited voices and a custom voice actor are Enterprise features.
- Model choice. Bland says it "owns the full voice AI stack", including the language model. That is why there are no token charges, and also why we found no option in its call API to pick an outside LLM.
- Self-hosting. On-prem and VPC deployment is available, but only on Enterprise.
Bland AI alternatives at a glance
| Tool | Price per minute, as stated | What the price includes | Free trial | Phone numbers | AI models you can choose |
|---|---|---|---|---|---|
| Bland AI | $0.14 (Start), $0.12 (Build, $299/month) | LLM, speech-to-text and voice; phone billed separately | Free Start plan, no card | 1 inbound number on Start; more $15/month; or your own Twilio or SIP | Bland's own stack; no outside LLM picker in its docs |
| Retell AI | $0.07 to $0.31 | $0.055 voice engine plus the LLM and voice you pick; phone extra ($0.015/min US on Retell's Twilio) | $10 in credits | $2/month; no charge on your own SIP trunk | GPT, Claude, Gemini, GPT Realtime, Grok Voice or your own LLM |
| Vapi | $0.05 platform fee; $82 to $129 per 1,000 minutes on its calculator | Hosting only; speech-to-text, LLM, voice and phone at provider cost | $5 in credits | 1 free Vapi number (5 on Core, $29/month) | Its full model catalog, at cost |
| ElevenLabs Agents | $0.08 per extra minute; plans $6 to $990/month include 75 to 12,375 minutes | Hosting, speech-to-text and voice; LLM billed on top; no telephony fee | Free plan, 15 minutes a month | Bring your own (Twilio or SIP trunk); the carrier bills you | Supported third-party LLMs, or your own model |
| Telnyx | $0.05 voice engine | Orchestration, speech-to-text and voice; LLM tokens and phone extra; billed in 60-second steps | Free sign-up | US local $1.00/month; calls from $0.0032/min | Open models on Telnyx GPUs, or Anthropic and OpenAI models |
| LiveKit Agents | $0.01 agent session; example $0.048 all-in | Hosting on LiveKit Cloud; models extra (or via LiveKit Inference); free to self-host | Free Build plan, no card | US local numbers; $0.01/min telephony in its example | Any model via plugins; Inference lists OpenAI, Google, xAI, DeepSeek and others |
| Pipecat | $0.01 to $0.03 hosting on Pipecat Cloud | Hosting only; models and transport extra; free to self-host | Not stated | Daily SIP $0.003 to $0.02/min, or PSTN | Any provider Pipecat integrates with |
| PolyAI Dialog API | 2 cents/min plus text tokens | The model only: no telephony or orchestration | $100 credit for sign-ups before November 19 | Not included in the API | PolyAI's own Dialog-RSN-1 |
| Phonely | Plans $50 and $150/month with minutes; then $0.25 or $0.30/min | Agent, premium voices, transcripts; Pro adds AI SMS and CRM integrations | Not stated | 3 free (Starter), 5 free (Pro); SIP trunking | Phonely's own Alma model |
| Synthflow | No per-minute rate: annual enterprise contract | Telephony, integrations, security review and launch support, per contract | None | Synthflow telephony, SIP or enterprise telephony | Not stated on the pricing page |
Picking by why you are leaving Bland
- A lower per-minute bill: Telnyx, at $0.05 plus LLM tokens (about $0.006 a minute on its own Kimi example) and carrier-cost telephony. Calls are rounded up to the next full minute, which matters if most of yours are short.
- Your choice of LLM: Retell AI lists GPT, Claude and Gemini models with a per-minute price for each, plus speech-to-speech options and custom LLMs. 20 concurrent calls are free (Bland's Start allows 10), and phone numbers are $2 a month. Vapi gives you the widest control at $0.05 plus providers at cost; see our Vapi alternatives guide.
- More voices without Enterprise: ElevenLabs Agents. Even the free plan includes 15 minutes and 4 concurrent calls; Business ($990 a month) adds 10 professional voice clones. Retell also lets you choose ElevenLabs, Cartesia, MiniMax, Fish, OpenAI or Inworld voices per agent.
- Self-hosting without an enterprise contract: LiveKit Agents (Apache-2.0) or Pipecat (BSD-2-Clause) are open-source frameworks you can run on your own servers with any model. Pipecat's maker Daily announced on October 29, 2026 that Deepgram is acquiring it, and said Pipecat will remain open source and vendor neutral.
- An enterprise vendor other than Bland: PolyAI and Parloa both sell per-minute platforms by quote. Synthflow is now enterprise-only too; see our Synthflow alternatives guide.
- A no-code agent for a small business: Phonely from $50 a month with minutes, numbers and unlimited concurrent calls included, or a done-for-you receptionist like Rosie from $49 a month for 250 minutes.
Bland is still the right pick for some teams: if you want one per-minute number that already includes the model, voice and transcription, and you are fine with Bland's own stack, its Start plan costs nothing to try.
¿Te está gustando? Recibe uno así en tu bandeja cada mañana.
un correo al día · date de baja en dos clics · sin rastreadores de terceros
Frequently asked questions
How much does Bland AI cost per minute?
On November 9, 2026, Bland's pricing page lists $0.14 a minute on the free Start plan and $0.12 on Build, which has a $299 monthly platform fee. Transferred-call time is $0.05 or $0.04 a minute. Enterprise pricing is custom.
Does Bland AI's price include the phone line?
Bland's pricing page says the per-minute rate covers the LLM, speech-to-text and voice, and that telephony is billed separately, on your own carrier or Bland's at pass-through cost. Phone numbers are $15 a month; the Start plan includes one inbound number.
Can you choose your own LLM on Bland AI?
Bland runs its own stack, including the language model, and we found no option in its call API to pick an outside LLM. If you want to choose GPT, Claude or Gemini, Retell AI, Vapi, ElevenLabs Agents and Telnyx all let you.
Can you self-host Bland AI?
Only on the Enterprise plan, which offers on-prem and VPC deployment. To self-host without an enterprise contract, use an open-source framework such as LiveKit Agents (Apache-2.0) or Pipecat (BSD-2-Clause).
What is the cheapest alternative to Bland AI?
On stated rates, Telnyx at $0.05 a minute including speech and voice, with LLM tokens and telephony extra (it estimates about $0.06 all-in), or LiveKit Cloud at $0.01 a minute for the agent session with an example all-in cost of $0.048. Self-hosting LiveKit or Pipecat removes the platform fee entirely.