Bland AI

Bland AI is a voice-first agent platform for phone conversations: it builds, tests, and runs AI agents that answer inbound calls, place outbound calls, and hand complex conversations to humans with the full transcript attached.

Bland positions the product for regulated, high-stakes calling - healthcare, insurance, financial services, and logistics - and runs agents on its own infrastructure so call data does not pass through third-party model providers.

On the homepage read on 20 September 2026 the company reports 672,759,720 calls resolved to date, support for 40+ languages, and a published latency figure of about 400 ms against an industry average of 1,240 ms.

Agents are assembled from conversational pathways, tools, and knowledge bases; the in-product builder (ask Norm) drafts the pathway, calendar action, and CRM step from a description, and every agent can be tested against a scenario suite before it touches production.

The marketing and revenue use cases are the ones that usually justify the spend.

Outbound agents qualify inbound leads, confirm appointments, and follow up on every account without adding headcount, while inbound agents greet callers, verify identity, answer questions, and route or book in a single conversation - effectively retiring the press-1 phone tree.

One agent can be deployed across voice, SMS, iMessage, and web chat with unified memory, so a caller who hangs up and texts in continues the same thread instead of starting over.

Supporting features include appointment scheduling and in-call SMS nodes, warm and live transfers, guardrails for protected calls, live call monitoring, and searchable transcripts that are transcribed, scored, and analysed so failures surface instead of hiding in recordings.

The pricing model is the clearest part of the pitch, because Bland publishes a rate card rather than asking for a demo.

The Start plan is $0.14 per minute with no platform fee and gives two credits plus an inbound number the company values at $15/month, no card required; Build drops to $0.12 per minute but adds a $299/month platform fee and raises concurrency from 10 to 50 concurrent calls with a 99.9% uptime SLA; Enterprise is quoted to your volume with on-premises or VPC deployment, a forward-deployed engineer, and BAAs, SSO, and data residency included in the per-minute rate.

Bland states that its per-minute rate already covers the LLM, speech-to-text, and text-to-speech with no token charges or model-provider pass-throughs, and that telephony is billed separately on your own carrier or on Bland's at pass-through cost - customers who bring their own telephony do not pay transfer fees.

For comparison, Bland's own pricing page notes that Vapi lists $0.05/min and Retell about $0.07/min for voice infrastructure alone, and estimates production stacks land around $0.13-$0.30 (Vapi) and $0.11-$0.25 (Retell) per minute once models and telephony are added; that is the vendor's published comparison, not our measurement.