Dust is live · free plan, paid from €24/seat/mo billed annually · last verified 4 Nov 2026 · viability healthy
Dust
Dust builds multiplayer AI agents that pull company context from 70+ connectors and act inside Slack, CRM and docs.
Dust is the pick when agents must read Salesforce, post to Slack and still clear a security review — SSO, Spaces, audit logs and US/EU residency all sit on the self-serve Business plan, which is unusual at this price. Budget carefully: Pro seats get 8,000 credits a month and unused credits do not roll over, so deep research or tool-heavy orchestration burns allocation fast. Without someone owning the AI Operator job, the workspace will sit idle.
Verdict generatedVerified 9d ago · liveness 75/100 · cite: rightaichoice.com/tools/dust
- Ops teams in support, sales, marketing or data wiring agents into real company systems like Slack, Salesforce and Zendesk
- Companies that need per-team data segmentation, SSO, SCIM and audit logs before a broad AI rollout
- Teams that want a different model per agent instead of one model everywhere
- Engineering and IT groups building incident response, ticket triage or self-service analytics agents
- Individuals who want a single personal chatbot with no setup — agents, skills and permissions need configuring
- Small non-technical teams with no ops owner to build, run and maintain agents
- Buyers who need flat-rate predictable billing — credit burn varies by model and tool use, and credits don't roll over
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Dust if you have no one to own agent configuration and credit budgeting — the platform only pays off once someone takes on the AI Operator role.
Credits reset at the start of every billing period and unused credits do not roll over, so a quiet month still costs the same.
The Business plan fits a 5–100 person team with one or two ops owners: Free seats at 0€ for occasional users, Pro at 24€/seat/mo billed yearly (30€ month-to-month) for most members, Max at 120€/seat/mo billed yearly (150€ month-to-month) for power users. It undercuts per-seat enterprise agent platforms while sitting above flat-rate automation tools like Zapier, because you pay for model consumption rather than just seats.
In short
Dust — Dust builds multiplayer AI agents that pull company context from 70+ connectors and act inside Slack, CRM and docs. Best for Ops teams in support, sales, marketing or data wiring agents into real company systems like Slack, Salesforce and Zendesk, Companies that need per-team data segmentation, SSO, SCIM and audit logs before a broad AI rollout, Teams that want a different model per agent instead of one model everywhere. Free to start; paid plans from €24/seat/mo billed annually.
What is Dust?
Dust — Dust builds multiplayer AI agents that pull company context from 70+ connectors and act inside Slack, CRM and docs. Best for Ops teams in support, sales, marketing or data wiring agents into real company systems like Slack, Salesforce and Zendesk, Companies that need per-team data segmentation, SSO, SCIM and audit logs before a broad AI rollout, Teams that want a different model per agent instead of one model everywhere. Free to start; paid plans from €24/seat/mo billed annually.
Dust is a platform for building and running AI agents that work from your company's actual systems rather than a blank chat window. You connect it to 70+ connectors — Slack, Notion, Google Drive, GitHub, Salesforce, Zendesk and more — then define agents with instructions, skills, knowledge and tools. Agents run from chat, on a schedule, or on an event trigger, and each one can use a different model from 20+ frontier and open-source options across OpenAI, Anthropic, Google, Mistral and DeepSeek. The company's framing is "multiplayer AI": people and agents share the same conversations, skills, tools and notifications, with human review built into the loop. Around that sit multi-agent orchestration, native and remote MCP servers, Pods (shared workspaces), Spaces for per-team data segmentation, and Frames for interactive dashboards and apps. A July 2026 release added self-improving Skills, reusable instruction sets that encode team expertise and adapt from usage — Dust says its own support team migrated from a single @support agent to @Dust powered by a Support skill. Governance is the part buyers actually evaluate. A dual-layer permission model separates what an agent can reach from who can run it, backed by SSO via Okta, Entra ID and Jumpcloud, SCIM provisioning, audit logs, usage reporting and US or EU data residency. SOC 2 Type II is published on the trust page; Enterprise adds single-tenant deployment, custom retention, workspace-pooled credits and custom legal terms. Dust reports 300,000+ agents deployed across 3,000+ teams, and raised $40M in May 2026 per Axios. It is aimed at ops-minded teams in support, sales, marketing, data and engineering that have someone willing to own the AI Operator role — Dust's term for the person who builds and runs agents for everyone else. Against lighter automation tools, Dust's edge is depth of company context plus the permission layer; against a single-model assistant, the edge is model choice per agent and a shared workspace
Key features
- Build custom agents with your own instructions, skills, knowledge and connected tools
- Choose a model per agent from 20+ frontier and open-source models (OpenAI, Anthropic, Google, Mistral, DeepSeek)
- Self-improving Skills: reusable instruction sets that encode team expertise and adapt from usage
- Multi-agent orchestration with scheduled and event-driven triggers
- 70+ connectors including Slack, Notion, Google Drive, GitHub, Salesforce and Zendesk
- Connect any other tool via native and remote MCP servers
- Frames: interactive dashboards and apps, white-labellable on Enterprise
- Pods: collaborative workspaces with shared context for people and agents
- Spaces for segmenting company data and permissions by team
- Dual-layer permission model separating agent data access from user access
- SSO via Okta, Entra ID and Jumpcloud, plus SCIM provisioning and deprovisioning
- Audit logs, usage analytics and adoption reporting
- US or EU data residency, with single-tenant deployment on Enterprise
- Developer API, Conversation API and Data Source API for programmatic access
- Forward an email thread to an agent to loop it into the conversation
Behind the verdict
The question we get most often is whether Dust is worth the setup cost versus just giving everyone ChatGPT or Claude. The answer hinges on one thing: do your agents need to touch systems that live behind SSO? If the work is "summarize this doc," you don't need Dust. If it's "triage new Zendesk tickets, check the account in Salesforce and route to the right queue," the connectors and the dual-layer permission model are doing real work that a chat window can't. We'd reach for it when a support, sales or GTM-engineering team already has a tinkerer. Clay and Vanta both put their GTM engineering leads on camera for Dust, which tells you the target buyer builds agents as part of the job rather than waiting on IT. Where it bites is the credit model. Pro is 8,000 credits per seat per month and Max is 40,000, both resetting to zero each billing period. Dust is upfront that a token-efficient model on basic chat burns few credits while a deep-research run burns a lot. A team that quietly starts running multi-step research agents will outgrow Pro seats and discover it in the invoice, not in advance. Overages exist only if an admin enables a capped workspace overage. Model flexibility is the other differentiator worth paying for. One agent can run a cheap model for classification and a heavier one for drafting, per workflow, and no model is locked behind a higher tier. That matters for cost control more than for benchmark scores. Compare it to n8n or Make if your primary need is deterministic automation: those are cheaper and more predictable, but they don't reason over unstructured company context. Compare it to a single-vendor assistant if you want zero configuration. Dust sits between those, and it wants an owner. The recent release cadence has leaned toward enablement and
Models under the hood
as of 2026-11-08
Limitations
- Dust is a platform rather than a plug-in: agents, skills, connectors and permissions all require real configuration, and someone has to own that work.
- The self-serve Business plan is scoped for teams up to 100 people, with unlimited connectors, MCP servers, workspace-pooled credits and volume pricing held behind Enterprise.
- Credit consumption is variable — it depends on the model chosen, task complexity and any tools the agent calls — so multi-step orchestration draws far more than basic chat, and each seat's monthly allocation resets rather than rolling over.
- Free seats get 500 credits for their lifetime.
as of 2026-10-11
Researching Dust? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Is Dust still active?
Dust is live · free plan, paid from €24/seat/mo billed annually · last verified 4 Nov 2026 · viability healthy.
Viability score
How well maintained and how widely used is Dust? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: November 2026
How we score →What's new in Dust
Checked 4 days agoAcross the latest 4 updates: 1 launch and 3 news mentions.
Dust welcomes AI Partners as an official partner
AI Partners, a Paris-based AI transformation firm, becomes an official Dust partner serving complex organizations.
Your AI deployment won't pay off: here's why
Dust argues most companies miscalculate AI ROI by counting saved hours instead of measuring deployment impact.
Introducing self-improving skills: reusable AI capabilities that learn from usage
Dust ships self-improving Skills — reusable instruction sets that encode team expertise and adapt from usage.
How Dust runs Support on Dust
Dust Support migrated from a single @support agent to @Dust, powered by a Support skill.
What do users say about Dust?
What people actually say about Dust — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
ResearchedDiscussed by users across 6 sources (Hacker News, YouTube, Product Hunt, App Store, Stack Overflow, Lemmy) · researched Sep 29, 2026.
Weighted by the 86 posts each of 6 sources contributed.
- +70+ native connectors (Slack, Notion, Salesforce, Zendesk, GitHub) let agents act on real company systems
- +Per-agent model choice across 20+ frontier and open-source models enables cost/quality tuning
- +Dual-layer permission model separates what agents access from who can use them
- +US or EU data residency plus SSO, SCIM and audit logs address enterprise governance needs
- +Multi-agent orchestration with scheduled and event-driven triggers supports real operational workflows
- −No authentic community feedback exists in the scraped data to validate the tool's real-world reliability
- −Credits reset each period with no rollover, risking mid-month exhaustion for busy teams
- −Enterprise-only features (unlimited connectors, pooled credits, single-tenant) leave smaller tiers constrained
- −Name collision with games, novels and a messaging app makes independent research genuinely hard
- −Credit-based pricing is opaque: translating 8,000 credits into real agent runs is non-obvious
- • No credit rollover — unused credits vanish each billing period
- • Credit consumption per agent run is not transparently documented
- • Pooled and unlimited connectors require Enterprise, forcing mid-market upgrades
How we verified this
Every listed fact about Dust — pricing, features, links, status — is re-read from the vendor's own pages by our verification pipeline and each change is logged with a date; the last pass ran on 4 Nov 2026. Analysis is drafted with AI models and checked against our editorial standards before publishing; see how we verify.
Verification history
We have re-verified Dust 12 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — changed pricing page
- — changed SSL certificate
- — changed pricing page, features, summary, description, our verdict, pricing tiers, who should skip it
- — changed SSL certificate
- — changed description, our verdict, summary, who it suits, features, pricing tiers
- — changed models, pricing page
Showing the 6 most recent of 12 verification passes.
Free to cite with attribution — this page re-verifies continuously.
How much does Dust cost in 2026?
Dust: Free to start; paid plans from €24/seat/mo billed annually. Pricing last verified 4 Nov 2026.
Plans compared
Every published Dust tier, its price as the vendor publishes it, who it fits, and what it adds over the tier below. The 12-month cost projection follows.
| Plan | Price | Billing | Who it's for |
|---|---|---|---|
| Free | FreeVendor: “0€” | — | Occasional users or teammates who just want to try Dust before the team commits to paid seats.Adds: Starting tier: 0€ with 500 credits for the seat's lifetime, plus SSO, 20+ models and connectors to 20+ data sources. |
| Pro | €24/seat/mo billed annuallyVendor: “24€/seat/mo billed yearly (30€ monthly)” | monthly, billed annually, per seat | Most team members running everyday agents — ticket triage, account briefs, report generation — several times a day.Adds: Adds 8,000 credits per seat per month, multi-agent orchestration, MCP servers, up to 3 Spaces and 5 remote Frames. |
| Max | €120/seat/mo billed annuallyVendor: “120€/seat/mo billed yearly (150€ monthly)” | monthly, billed annually, per seat | Power users running deep research, complex automations or tool-heavy multi-step orchestration every week.Adds: Same feature set as Pro with 40,000 credits per seat per month instead of 8,000 — a 5x allocation for heavier workflows. |
| Enterprise | Custom | quote | Companies past 100 people, or anyone who needs single-tenant deployment, custom retention or white-labelled Frames.Adds: Adds unlimited connectors and MCP servers, workspace-pooled credits, volume pricing, SCIM, audit logs, custom legal terms and a dedicated CSM with SLA. |
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Where the pricing makes sense
The company stage and team size where Dust's pricing actually pencils out — and where peers do it cheaper.
The Business plan fits a 5–100 person team with one or two ops owners: Free seats at 0€ for occasional users, Pro at 24€/seat/mo billed yearly (30€ month-to-month) for most members, Max at 120€/seat/mo billed yearly (150€ month-to-month) for power users. It undercuts per-seat enterprise agent platforms while sitting above flat-rate automation tools like Zapier, because you pay for model consumption rather than just seats.
Has Dust's pricing changed?
No plan price for Dust has moved since we began logging in Aug 2026; the pricing page is re-read on every verification pass.
Who should pick Dust?
Pick Dust if you are Ops teams in support, sales, marketing or data wiring agents into real company systems like Slack, Salesforce and Zendesk, Companies that need per-team data segmentation, SSO, SCIM and audit logs before a broad AI rollout, Teams that want a different model per agent instead of one model everywhere, Engineering and IT groups building incident response, ticket triage or self-service analytics agents.
Look elsewhere if you are Individuals who want a single personal chatbot with no setup — agents, skills and permissions need configuring, Small non-technical teams with no ops owner to build, run and maintain agents, Buyers who need flat-rate predictable billing — credit burn varies by model and tool use, and credits don't roll over, Groups expecting heavy quarterly platform churn — recent Dust news skews toward partners and enablement content.
Skip it if Skip Dust if you have no one to own agent configuration and credit budgeting — the platform only pays off once someone takes on the AI Operator role.
GeneratedReal-world workflow fit
Concrete scenarios for the personas Dust actually fits — and what changes day-one when you adopt it.
Connects Zendesk and Notion, builds a triage agent on a cheap model to classify and route incoming tickets, then a second agent on a heavier model to draft replies grounded in the knowledge base. Escalation detection surfaces at-risk accounts in Slack.
Outcome: Tickets route instantly instead of sitting in a queue, and resolved tickets get recycled into reusable internal docs rather than disappearing.
Wires Salesforce, Slack and Google Drive into an account research agent, assigns Pro seats to reps and Max seats to the two people running deep research workflows, and monitors credit burn in the usage dashboard.
Outcome: Reps get account briefs in Slack before calls, and the workspace stays inside a budgeted credit ceiling the admin controls.
Deploys on the Enterprise tier with EU residency, SCIM provisioning from Entra ID, per-team Spaces and audit logs, then lets analysts ask pipeline questions in plain English through a Frame.
Outcome: Security signs off on the rollout because permissions and audit trails exist up front rather than as a later add-on.
Use cases
- Automate weekly business report generation from your CRM data
- Build a support agent that answers from your knowledge base and escalates at-risk accounts
- Classify and route incoming support tickets to the right team instantly
- Brief sales reps in seconds with the latest signal on an account
- Score and route inbound leads against your live ICP
- Turn resolved tickets into reusable internal documentation
- Watch dashboards and alert on pipeline anomalies automatically
- Let teammates ask data questions in plain English and get answers from connected pipelines
Setup time & first value
How long it actually takes to get something useful out of Dust — broken out by persona, not the marketing-page minute.
A single connector plus one simple agent is roughly an afternoon for someone comfortable with APIs. A department-wide rollout with SSO, SCIM, Spaces and permission mapping is a multi-week project with a named owner. Free-tier evaluators can try an agent in under an hour, but Free seats cap at 500 lifetime credits.
Switching to or from Dust
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From a single-model chat assistant: move your prompt library into agents and Skills, then attach the connectors the chat tool couldn't reach.
- →From Zapier or Make automations: rebuild the multi-step flows as Dust agents on triggers, and keep Zapier, Make or n8n as connected tools where they still do the job.
- →From internal scripts and cron jobs: replace scheduled data extraction with Dust agents on the same schedule, gaining audit logs and per-team permissions.
- →From a per-department point tool: consolidate into Pods so people and agents share one context instead of separate knowledge bases.
- ↗To a flat-rate automation platform: export agent logic back into trigger-based workflows, accepting the loss of per-agent model choice and audit logs.
- ↗To a suite-native assistant: rebuild agents inside your existing productivity suite, losing model flexibility and cross-tool context.
- ↗To an in-house build: re-implement orchestration against provider APIs directly, taking on permissions, metering and audit trails yourself.
Integrations
Resources & guides
Tutorials & Learning
YouTube returned 6 videos for “Dust”, and we withheld 6: 6 could not be judged, because “Dust” is a single word that other videos use for other things. We are showing none, because we could not prove any of them are about Dust.
Official links
What are the best Dust alternatives?
The closest alternatives to Dust we list are Zapier, Gumloop and Coze. Ranked by direct product-type match, not category overlap.
| Alternative | Pricing | Starts at | Last verified |
|---|---|---|---|
| Zapier | Freemium | Pricing not published | — |
| Gumloop | Freemium | Pricing not published | — |
| Coze | Freemium | Pricing not published | — |
Alternatives to Dust
View all
Tools that pair well with Dust
Common stack mates teams adopt alongside Dust, with the specific reason each pairing earns its keep.
Zapier
Zapier connects 9,000+ apps to Claude, ChatGPT, Cursor and Gemini Enterprise through one governed MCP server.
Gumloop
Build AI agents for work: visual canvas, Slack @mentions, and a real browser for every agent
Coze
Coze lets you build no-code AI agents and publish them as web apps, APIs, and chat bots across Discord, Telegram, or Slack.
Dust vs its closest rival
Dust's closest head-to-head is with Truleo — These two are not competitors and no buyer should be picking between them. Truleo is vertical software for law enforcement: it connects to a department's existing RMS, CAD, LPR, BWC and jail-call systems and returns ranked case solvability scores, leads and briefings — priced per connected application plus seats, with a 60-day no-card trial. Dust is a horizontal agent platform for company ops teams that want custom agents over Slack, Notion, Salesforce and 70+ other connectors, with per-agent model choice from 20+ options. If you run a police agency, evaluate Truleo on integration coverage and data-handling policy. If you run an ops function, evaluate Dust on connectors, permissions and credit economics. Comparing them head-to-head is apples-to-oranges. Verdict generated
Head-to-head comparisons
Dust vs Truleo
These two are not competitors and no buyer should be picking between them. Truleo is vertical software for law enforcement: it connects to a department's existing RMS, CAD, LPR, BWC and jail-call systems and returns ranked case solvability scores, leads and briefings — priced per connected application plus seats, with a 60-day no-card trial. Dust is a horizontal agent platform for company ops teams that want custom agents over Slack, Notion, Salesforce and 70+ other connectors, with per-agent model choice from 20+ options. If you run a police agency, evaluate Truleo on integration coverage and data-handling policy. If you run an ops function, evaluate Dust on connectors, permissions and credit economics. Comparing them head-to-head is apples-to-oranges.
GeneratedDust vs Presto Voice
These aren't competitors — nobody shortlists both. If you're an ops, support, or data team trying to wire AI into Slack, Salesforce, and Zendesk, Dust is the only one of the two that applies. If you run drive-thrus for a QSR franchise group, Presto Voice is the only one that applies, and the Sept 21, 2026 Toast Partner Ecosystem listing is the shortest integration path if you're on Toast POS. The honest answer is that your problem, not this page, decides which one you're even evaluating.
GeneratedDust vs Locus Robotics
These are not competitors, and treating them as a head-to-head would be a category error. Dust sells software agents that sit inside Slack, Salesforce, Zendesk and 70+ other tools so ops teams can automate knowledge work. Locus Robotics sells robots and a RaaS subscription that physically move inventory across a warehouse floor. If your problem is 'our support and sales ops teams need agents wired into company systems,' Dust is the relevant evaluation. If your problem is 'our fulfillment center needs 2–3x picking productivity without rebuilding the racking,' Locus is the relevant evaluation. A buyer with both problems buys both — they don't choose between them.
GeneratedDust vs Writer
Pick Dust if your problem is wiring agents into the tools your company already runs on — Slack, GitHub, Salesforce, Zendesk — and you want per-agent model choice plus segmenting company data by team with SSO and SCIM on top. Pick Writer if your problem is getting hundreds of people to produce work that sounds like your brand and survives audit — especially in finance or healthcare — and you'll trade model flexibility for governed, brand-encoded agents on Writer's own Palmyra X6. Neither is a fit if you want a personal no-setup chatbot: both require real configuration and an owner to maintain them.
GeneratedDust vs Toolhouse
Verdict being re-reviewed against current pricing and data.
Dust vs Genspark
Pick Dust if agents need to act inside your company's real systems — Slack, Salesforce, Zendesk, GitHub — and you have an ops owner who will configure permissions, Spaces and SSO. Pick Genspark if the job is know-how work: cited research, decks, docs, podcasts and no-code internal tools, ideally on Google Workspace or Microsoft 365. If you cannot name the systems the agent must touch, Dust's setup cost is wasted and Genspark is the faster win.
GeneratedDust vs Cryptohopper
These aren't competitors, so there's no head-to-head pick to make. Dust is for ops, support, sales, and data teams that want agents wired into Slack, Notion, Salesforce, or Zendesk with per-team data segmentation, SSO/SCIM, and audit logs — and it's freemium with consumption-based credits that don't roll over. Cryptohopper is a paid crypto trading bot: Explorer starts at $24.16/mo after a 3-day trial, and you're buying Copy Bot, DCA, trailing orders, backtesting, and a no-code Strategy Designer across exchanges like Binance and Kraken. Buy Dust to deploy multiplayer AI inside your company; buy Cryptohopper to automate crypto trades.
GeneratedDust vs Air Ai
These are not competitors — this is not a head-to-head anyone should run. If you're a commercial ops, support, or engineering team that wants to build agents on top of Slack, Notion, Salesforce, and GitHub with per-team data segmentation and SSO, Dust is the relevant product and Air is simply out of scope. Air is for military commands, acquisition and sustainment offices, and defense program teams with a readiness mission and existing enterprise systems to feed its Readiness Graph; it's sold through a vendor-led deployment, with no published price and no self-serve path. No buyer with one budget and one problem shortlists both. Choose based on which world you're in, not on feature-by-feature scoring.
GeneratedDust vs Notable
These are not competitors, so pick by what you are. If you run ops in a non-healthcare company and need a customizable agent layered onto Slack, Drive, Salesforce and the rest, Dust is the only one of the two that fits — freemium entry, credit-based cost, 70+ connectors, per-agent model choice. If you are a health system trying to cut prior-auth backlog, denial write-offs and call volume, Notable is the specialist with the payer plumbing (Epic, Cerner, athenahealth, Twilio) Dust simply does not have. Nobody should be comparing these side by side; they solve different problems for different buyers.
GeneratedFrequently asked questions
Categories
Best-of guides
For makers of Dust
- Get a Verified Listing for Dust ($39): a written decision within 1 working day.
- Free listing for another tool: Free reviews are decided in order, within 3 weeks of the review starting.
Used Dust? Help shape our editorial sentiment research.