Chatbase
Builder for docs-grounded RAG chatbots with actions.
Chatbase is a no-code platform for building and deploying custom AI support agents trained on a company's own data. Founded in 2023 and bootstrapped to serve thousands of businesses, it was one of the early chat-with-your-data products, and its pitch is letting a non-technical team stand up a ChatGPT-style agent for their business in minutes rather than building a custom stack.
The starting point is your knowledge. You feed Chatbase website URLs, which it crawls, along with PDFs, documents, spreadsheets, raw text, Q&A pairs, or a Notion workspace, and it can even ingest existing Zendesk and Salesforce support tickets as training data. From that material it builds a conversational agent that answers customer questions grounded in your content.
What pushes Chatbase past a simple FAQ bot is AI Actions, which let the agent take real steps in external systems. Connected to tools like order management, a CRM, or a helpdesk, an agent can look up an order, update a subscription, change a customer's address, capture a lead, or book a meeting, with integrations including Stripe, Shopify, Calendly, Slack, Zendesk, and Salesforce. When a query needs a person, the agent escalates to a human via live chat or a helpdesk ticket based on instructions you write in plain language.
A distinctive strength is model flexibility. Rather than locking you into one provider, Chatbase offers many leading models from several providers, including OpenAI, Anthropic, Google, and xAI, and lets you switch models or compare them side by side in a Playground where you also set the system prompt, configure actions, and test responses live.
Agents deploy across a website widget, WhatsApp, Messenger, Instagram, Slack, and email, with chat localization across many languages. Chatbase adds analytics on resolutions and engagement so the agent improves over time, a public API for deeper integration, and enterprise controls like SSO, encryption, and compliance, fitting teams that want a hosted product rather than a maintained custom runtime.
Vendor details
Canonical URL
https://chatbase.co
Category
Customer support agent
Subcategory
Support — chatbot builder
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
In practice
You want a support agent live this week, not a six-month engineering project. Chatbase trains on your website, PDFs, and help docs, and you deploy a working agent on your site, WhatsApp, or Slack in minutes.
Your bot can answer questions but can't actually do anything. Chatbase's AI Actions connect it to your order system and CRM, so it looks up orders, updates subscriptions, and changes addresses, not just replies.
You don't want to bet your support on one AI provider. Chatbase lets you pick from models across OpenAI, Anthropic, Google, and xAI, and compare them side by side in the Playground before you commit.
Agentic Index coverage score
10.0 / 14 capabilities · 71%
| Integrations & Tool CallingOut of the box integrations across Zendesk, Slack, Zapier, Stripe, WhatsApp and Messenger plus Shopify, Salesforce, Calendly and Cal.com, with real time connectors syncing to CRMs, order systems and help desks so replies reflect live business records rather than stale snapshots, Chatbase platform documentation 2026-07-30 | Full |
|---|---|
| Workflow OrchestrationAI Actions execute transactional workflows end to end rather than answering questions, including managing orders through Shopify, creating Zendesk tickets with drafted responses, scheduling through Calendly and Cal.com, and handing off to live agents via Salesforce Omni Channel with full conversation context carried across, Chatbase platform documentation and product blog 2026-07-30 | Full |
| Knowledge Grounding & RAGIngests documents, web pages, help center articles and raw text into a searchable knowledge base retrieved at query time by a RAG architecture, can additionally ingest Zendesk and Salesforce support tickets as training data, and Source Suggestions automatically identify knowledge gaps from live conversations, Chatbase platform documentation 2026-07-30 | Full |
| Human Oversight & GuardrailsA prompt architecture holds the model inside the configured knowledge scope so the agent refuses to answer outside its domain, an Improve Answer correction loop lets operators fix mistakes that then persist, live agent handoff carries full context, and RBAC with custom roles spans 14 permission areas; Chatbase itself states no guardrail system is perfect and hallucination prevention is probabilistic, Chatbase product blog and security page 2026-07-30 | Full |
| Security, Identity & GovernanceSOC 2 Type II and GDPR compliant with AES 256 encryption at rest and in transit, SSO on the Enterprise plan added January 2026, RBAC with custom roles across 14 permission areas, per user rate limiting, domain allowlisting controlling where the agent can be embedded, and customer data never used to train models, Chatbase security page 2026-07-30 | Full |
| Observability & AuditabilityAnalytics cover topic clustering, sentiment analysis, confidence scoring and geographic distribution with export to JSON, PDF and CSV, alongside reviewable chat logs, but this is conversation performance analytics rather than agent decision tracing or a configuration audit trail, Chatbase product blog 2026-07-30 | Partial |
| Memory & State PersistenceA Contacts store and persisted chat logs hold customer records and history, and full conversation context transfers on human handoff, but no cross session agent memory layer was documented on the pages reviewed, Chatbase platform documentation 2026-07-30 | Partial |
| Deployment & Data ResidencyDelivered purely as multi tenant SaaS with no deployment model choice, region selection or data residency commitment documented on the security or platform pages reviewed; terms of service and DPA pages were not retrieved and are the re verification target for this cell, Chatbase security page 2026-07-30 | Unable to verify |
| Prebuilt Agents, Templates & PacksPre built Actions ship for the common support and commerce jobs and agents can be duplicated as a starting point, but no agent template library, marketplace or named prebuilt agent catalogue was retrieved on the pages reviewed, Chatbase platform documentation 2026-07-30 | Partial |
| Triggers & Channel CoverageOne agent configuration deploys natively to six channels, website widget, WhatsApp, Messenger, Instagram, Slack and email, plus a hosted Agent Page and full API access, with voice input added across all channels and chat UI localisation in 40 plus languages including right to left support, Chatbase platform documentation 2026-07-30 | Full |
| Model Flexibility & RoutingModel comparison and selection is named as a platform differentiator, with the no code builder letting the operator choose and directly compare LLM models for an agent rather than accepting a fixed provider, Chatbase platform documentation 2026-07-30 | Full |
| APIs, SDKs & MCP ExtensibilityPublic developer documentation describes a REST API for chatting with agents, managing agents programmatically, and retrieving conversations, leads and analytics, with full API access listed alongside the native channel deployments; capability availability varies by pricing tier, Chatbase developer documentation 2026-07-30 | Full |
| Testing, Debugging & OptimizationConfidence scoring flags weak answers and the Improve Answer feature lets operators correct mistakes over time from reviewed chat logs, but no pre deployment test environment, simulation harness or scored evaluation suite was documented, Chatbase platform documentation 2026-07-30 | Partial |
| Browser & Computer UseNo browser control or computer use capability documented; the agent acts through configured AI Actions against connected APIs and its six native channels rather than by operating a browser or a human interface, Chatbase platform documentation 2026-07-30 | Unable to verify |
The Agentic Index coverage score grades every vendor Full, Partial or Unable to verify against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Recent platform changes
Chatbase introduced full API capabilities allowing users to create, configure, train, clone, and delete AI agents programmatically. The update includes endpoints for automatic agent creation via website crawling and modifying settings like AI models, instructions, and rate limits.
Bears on: MCP / tool calling / API
View sourcePricing
From $40/mo · free tier
credits
Included quota
Hobby (entry paid, $32 annual / $40 monthly): 500 message credits/mo, 1 agent with advanced models, 5 AI Actions, 10MB training, 2 seats, API, integrations, basic analytics. Free: 50 credits/mo, 1 agent, 400KB, agents deleted after 14 days idle, watermark, no API/Actions.
What is public
Chatbase (AI chatbot builder, ~10K+ paying customers) is credit-metered across five tiers (annual / monthly; annual ~20% off): Free ($0, 50 credits/mo, 1 agent, agents deleted after 14 days idle), Hobby $32/$40 (500 credits, 5 AI Actions, API), Standard $120/$150 (4,000 credits, 2 agents, voice, outbound), Pro $400/$500 (15,000 credits, 5 agents, advanced analytics), and custom Enterprise (SLA, CSM, white-label). Credit allotments were cut in 2026 (Hobby was 1,500, Standard 10,000, Pro 40,000).
Billing mechanics
Every AI response consumes message credits, weighted by model: standard models 1 credit, GPT-5.2 / Gemini Pro 2, Claude Sonnet 4.6 / Grok 3 = 3, Grok 4 = 4, Claude Opus 4.6 = 5. So a 500-credit Hobby plan is ~500 responses on cheap models but only ~100 on Opus. Credits reset monthly on the 1st; run out and the widget shows 'unavailable' unless auto-recharge is on. Tiers also gate agents, AI Actions, training-data size (KB-MB), and seats.
Cost watchouts
Model-weighted credit burn (premium models 3-5x faster), auto-recharge ballooning bills on erratic volume, branding removal ($39/mo+, up to ~$99-$199 by tier), custom domain add-on (~$59/mo / Enterprise), extra agents ($7/mo each), 2026 credit-allotment cuts, free agents deleted after 14 days idle
Variable cost rationale
Pure credit metering with model-weighted consumption and auto-recharge makes spend hard to predict - a viral traffic spike or a premium-model config can multiply the bill fast
Additional watchouts
The credit model plus add-ons (auto-recharge, branding removal, extra agents) routinely pushes real cost well above the sticker; Free/Hobby are undersized for production; no visual flow builder (LLM-driven only)
Overage / add-ons
Auto-recharge tops up credits when your balance drops below a threshold (~$40 per 1,000 non-expiring credits; manual add-on packs also ~$12-$14/1,000); without it, the bot stops responding at zero. Extra AI agents ~$7/agent/mo; premium-model selection silently multiplies credit burn.
Sales call required
No, self serve available
Free / trial
Free tier
Lowest paid plan
Free; up to ~$500/mo
Commercial notes
Train agents on websites/PDFs/Notion; AI Actions (Stripe/Calendly/Zendesk/Salesforce, order/cart ops); Shopify app since Jan 2026; model-agnostic (OpenAI/Anthropic/Gemini/DeepSeek/Grok/etc.); GDPR-compliant, no training on your data; AWS US hosting; ~$8M+ ARR
Key ambiguities
Credit allotments and add-on rates have changed in 2026 (sources cite both old and new numbers); real capacity depends entirely on model choice; branding-removal/custom-domain pricing varies by source/tier
Cancellation / refund
Self-serve monthly or annual (~20% off / 2 months free); credits reset monthly (don't roll over, except non-expiring auto-recharge top-ups); Enterprise on custom terms
Support SLA / resale
Pro adds priority support and advanced analytics; Enterprise adds SLAs, a dedicated CSM, white-label, custom domain, and higher limits; support responsiveness is a noted weak point in reviews
Missing data
Current credit allotments differ across sources after 2026 cuts; exact add-on rates (branding removal, custom domain, recharge) vary; Enterprise pricing is custom. Note: the prior seed 'From $500/mo' anchored on Pro's monthly price; entry corrected to Hobby (~$40/mo monthly, $32 annual).
Related vendors
- Decagon — AI agent platform purpose-built for customer support, with deep…
- 5.Y — Early stage Singapore platform (GLUCOSE) deploying goal driven AI…
- Ada — Agentic customer experience platform whose AI agents autonomously…
- Agency AI — Agentic AI for customer success from Drift co-founder Elias Torres,…
- Aisera — Enterprise agentic AI platform that orchestrates specialized agents…
- ASAPP — Generative AI for enterprise contact centers, with human-in-the-loop…
Alternatives to Chatbase
The closest documented capability profiles to Chatbase among customer support agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Botpress11.0 / 14Adds documented Deployment & Data ResidencyChatbase vs Botpress →
- Kommunicate9.0 / 14A lighter documented profile than ChatbaseChatbase vs Kommunicate →
- Ada11.0 / 14Fuller documented coverage on Observability & Auditability and Prebuilt Agents, Templates & Packs
- Aisera11.0 / 14Adds documented Deployment & Data Residency
- ASAPP11.0 / 14Fuller documented coverage on Observability & Auditability and Memory & State Persistence
- Assembled10.0 / 14Fuller documented coverage on Observability & Auditability and Prebuilt Agents, Templates & Packs
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded