Dust
by Dust
Custom AI assistants connected to company data sources such as Notion, Slack, Google Drive, and GitHub. Dust is an enterprise AI assistant platform founded in 2023 in Paris, France. The product allows non-developer teams to build custom assistants that retrieve answers from connected data sources rather than relying solely on a base model.
Performance Scores
2 rankings evaluated
Score range: 7.3 – 7.3
-
#2Best AI Agent Builders for Non-Developers in 2026
Score: 7.3 · Best for: Mid-market and enterprise teams building internal Slack-based knowledge assistants
-
#6Best LLM App Platforms for Building AI Agents in 2026
Score: 7.3 · Best for: Companies that want a shared AI workspace connected to internal documents and tools, with workspace-level governance.
Key Facts
Pricing
| Attribute | Value | As of | Source |
|---|---|---|---|
| Dust pricing (July 2026) | Dust pricing (verified July 2026) is credits-metered per seat: Free (14-day Pro trial), Pro $30/seat/mo ($24 annual; 8,000 credits/seat/mo), a new Max seat $150/seat/mo ($120 annual; 40,000 credits/seat/mo), and custom Enterprise. Connectors (Notion, Slack, Drive, GitHub, M365) and a choice of Claude/GPT/Mistral models are included, with credit overages billed separately. | Jul 2026 | Dust pricing |
General
| Attribute | Value | As of | Source |
|---|---|---|---|
| Pricing | Credit-metered since mid-2026: Pro $30/seat/mo ($24 annual, 8,000 credits/mo), Max $150 ($120 annual, 40,000 credits), Free seat with 500 lifetime credits; Enterprise custom | Jul 2026 | Dust pricing page |
| Connectors | Connectors to 20+ data sources (Slack, Notion, GitHub, Google Drive, etc.) plus any tool via MCP servers, as of July 2026 | Jul 2026 | Dust pricing page |
| Founded | 2023 | Apr 2026 | Sequoia Capital |
| Key Differentiator | Founded by Stanislas Polu (ex-OpenAI) and Gabriel Hubert; $40M Series B co-led by Sequoia (May 2026, $60M+ total raised); permission-aware AI agents over connected company data | Jul 2026 | Dust press release (GlobeNewswire) |
Strengths
- ●Native ingestion connectors for Notion, Drive, Slack, GitHub, and Intercom
- ●Agents can be deployed inside Slack as chat-native assistants for employees
- ●Workspace-grade access control and per-source permissioning
- ●Strong fit for internal knowledge assistants and on-boarding bots
- ●Strong native connectors to internal knowledge sources
- ●Workspace governance with audit logs and per-assistant permissions
- ●Suits horizontal company-wide deployment more than narrow product use cases
Limitations
- ●Less suited to outbound or always-on agents that act on email and calendar events
- ●Visual workflow building is shallower than Gumloop's
- ●Pricing is credit-based and charged per seat, so cost grows with team size: Pro is 24 EUR per seat per month on yearly billing with 8,000 credits, and Max is 120 EUR with 40,000 credits (dust.tt/pricing, read 29 September 2026)
- ●Builder is lighter than dedicated agent or workflow platforms
- ●Targets internal company use — less suited for customer-facing product agents
- ●Fewer model-routing controls than Vellum or CrewAI
Based on evaluations in 2 rankings: Best AI Agent Builders for Non-Developers in 2026, Best LLM App Platforms for Building AI Agents in 2026
Pricing Plans
Free seat
Free seat within the Business plan; 500 lifetime credits (no monthly refresh)
- ✓20+ frontier models (GPT, Claude, Gemini, Mistral, DeepSeek)
- ✓Custom agents with skills, knowledge and tools
- ✓Standard connectors (Slack, Notion, GitHub, Drive)
- !500 credits lifetime — does not reset monthly
Pro seat
($24.00/mo billed annually)
$30/seat/mo month-to-month, $24/seat/mo billed yearly; 8,000 credits/seat/mo (no rollover)
- ✓20+ frontier models (GPT, Claude, Gemini, Mistral, DeepSeek)
- ✓Connect Slack, Notion, GitHub, Drive + 20 more, or any tool via MCP
- ✓Multi-agent workflows on schedules and triggers
- ✓SSO with Okta, Entra ID and Jumpcloud
- !8,000 credits/seat/mo; unused credits do not roll over
- !Programmatic usage billed at $0.01/credit
Max seat
($120.00/mo billed annually)
$150/seat/mo month-to-month, $120/seat/mo billed yearly; 40,000 credits/seat/mo (no rollover)
- ✓Everything in Pro seat
- ✓5x credit allowance for power users
- ✓Suited to Deep Research and tool-heavy multi-step workflows
- !40,000 credits/seat/mo; unused credits do not roll over
Enterprise
Custom pricing (talk to sales); workspace-pooled credits and volume pricing
- ✓Unlimited connectors and MCP servers
- ✓SCIM, audit logs and custom data retention
- ✓Single-tenant deployment, US data residency
- ✓Dedicated CSM, priority support, SLA, custom legal terms (MSA/DPA)
About Dust
Dust is an enterprise AI assistant platform founded in 2023 in Paris, France. The product allows non-developer teams to build custom assistants that retrieve answers from connected data sources rather than relying solely on a base model.
Dust ships with managed connectors for Notion, Slack, Google Drive, GitHub, Intercom, Confluence, and Microsoft 365. Documents are indexed and made available to assistants through retrieval-augmented generation, with permissions inherited from the source system. Builders compose assistants from instructions, model choice (Claude, GPT-4 family, Mistral), tools, and connected data spaces. The platform also offers Dust Apps for custom code-based pipelines. Pricing is credit-metered since mid-2026: Pro at $30/seat/month ($24 annual, 8,000 credits/month), Max at $150/seat/month ($120 annual, 40,000 credits), and Enterprise custom-quoted. Source-level permissions and SSO are available on Enterprise.
Integrations (4)
Other AI Agent Platforms Tools
CrewAI
Open-source Python framework for building and orchestrating multi-agent AI systems
AI Agent PlatformsGumloop
No-code AI workflow automation with visual node-based editor
AI Agent PlatformsLangflow
Visual low-code platform for building AI agents and RAG applications with drag-and-drop components
AI Agent PlatformsLindy
AI agent platform for building autonomous digital workers
AI Agent PlatformsSee How It Ranks
Most Customizable AI Agent Platforms 2026
Most coverage of AI agent platforms rewards how quickly a working agent can be stood up. This ranking asks the opposite question: once it exists, how far can it be bent? What can be plugged into it, how much of the logic sits under the builder's control, and how much of that…
Best AI Agent Builders for Non-Developers in 2026
Eight AI agent builders are ranked here for operations, marketing and customer-success teams that construct multi-step AI agents without writing production code. Gumloop scores 8.1 and ranks first, followed by Dust at 7.3, Relevance AI 7.2, Lindy 7.0, Replit Agent 6.2, Sema4.ai…
Questions About Dust
What is the best LLM app platform in 2026?
As of April 2026, the leading LLM app platforms are LangChain (most-used Python and JS framework), Vellum (production prompt and eval platform), Langflow (open-source visual builder), Dust (workspace assistants), and LlamaIndex (data-framework for RAG). Choice depends on visual…
What is the best AI agent platform in 2026?
As of April 2026, the leading AI agent platforms are Relevance AI (low-code AI workforce), Lindy (personal AI assistants), Dust (workspace-grade assistants), Gumloop (visual AI workflows), and Relay.app (human-in-the-loop automations; shutting down September 2026). Choice depends…
How do you build a Dust assistant in 2026?
As of April 2026, building a Dust assistant takes four steps: connect data sources (Notion, Slack, Drive, GitHub, Salesforce), define the assistant's instructions and model, attach the relevant data sources or knowledge bases, and publish to a workspace channel. Assistants run…
What are the best Dust alternatives in 2026?
As of April 2026, the leading Dust alternatives are Relevance AI (low-code AI workforce), Lindy (personal AI assistants), Gumloop (visual AI workflows), Relay.app (human-in-the-loop automations; shutting down September 2026), and Glean (enterprise search and assistants). Choice…