Amazon Q is AWS's family of generative-AI assistants. It splits into two distinct products that share a name and almost nothing else: Q Developer, an AI coding assistant for engineers, and Q Business, an enterprise assistant that answers questions over your company's own data. This guide disambiguates the two, maps every surface Q runs in, lays out pricing for both, and benchmarks Q against GitHub Copilot, ChatGPT Enterprise, and Microsoft 365 Copilot.
Amazon Q is AWS's brand for generative-AI assistants. The name covers two products with different buyers, different surfaces, and different pricing — which is the single most common source of confusion about Q.
Amazon Q launched in late 2023 and reached general availability through 2024. AWS positions it as "the most capable generative-AI assistant for accelerating software development and leveraging your enterprise data." That one sentence already contains both products: "accelerating software development" is Amazon Q Developer, and "leveraging your enterprise data" is Amazon Q Business.
Under the hood, Amazon Q is built on Amazon Bedrock — AWS's managed foundation-model service — so it inherits Bedrock's security and data-handling guarantees. You do not pick the underlying model the way you do when you call Bedrock directly; AWS manages model selection and routing inside Q. What you choose is which product you are buying and which tier.
The practical takeaway: when someone says "we're looking at Amazon Q," the first question is always "Developer or Business?" An engineering leader evaluating an AI pair-programmer wants Q Developer. A CIO who wants employees to ask "what's our refund policy?" and get a cited answer from internal docs wants Q Business. Conflating them leads to the wrong pilot, the wrong budget line, and the wrong success metric.
There is also a layer of Q embedded inside other AWS services — QuickSight, Connect, and others — where "Q" appears as a capability rather than a standalone product you license per seat. We map all of these surfaces in section IV so the full footprint is clear.
Q Developer = AI coding assistant for engineers (IDE, CLI, console). Q Business = enterprise assistant that answers questions over your own data (RAG, ~40+ connectors). Same brand, same Bedrock-backed security posture, completely different jobs.
These are the two products you actually license. Read this section once and the rest of the Amazon Q landscape falls into place.
The two products diverge on almost every axis: who the buyer is, where it runs, what "good" looks like, and how it's priced. The only things they share are the brand "Amazon Q," the Bedrock foundation underneath, and the enterprise data-handling guarantees (your data is not used to train the base models).
Buyer: engineering leaders, platform teams, individual developers.
Where it runs: inside the IDE (VS Code, JetBrains IDEs, Visual Studio, Eclipse), the command line, and the AWS Management Console as an inline help and troubleshooting agent.
What it does: inline code completion across 15+ languages, a chat panel for "how do I…" questions, agents that can implement a multi-file feature from a prompt (/dev), generate unit tests and documentation, run guided code transformations and language/framework upgrades (/transform), and scan code for security vulnerabilities. It is AWS-aware — it can answer questions about your AWS account and reason about your architecture.
Success metric: accepted-suggestion rate, time-to-merge, reduction in boilerplate and upgrade toil.
For full feature depth, IDE setup, and the per-tier breakdown, see the dedicated Amazon Q Developer guide.
Buyer: CIOs, IT, knowledge-management and internal-operations teams.
Where it runs: a web experience your employees log into (often via single sign-on), an embeddable chat widget, browser extensions, and a Slack/Teams integration. It is not an IDE tool.
What it does: connects to your enterprise content — Microsoft SharePoint, OneDrive, Salesforce, ServiceNow, Confluence, Jira, Google Drive, Amazon S3, and roughly forty-plus other sources — indexes it, and answers natural-language questions with citations back to the source document. It respects each user's existing permissions, so an employee only ever gets answers from documents they were already allowed to read. It can also summarize, draft, and (with Q Apps) let employees build small no-code internal apps on top of that data.
Success metric: deflected internal support tickets, time-to-answer for employees, adoption across departments.
For connector lists, the index sizing model, permission handling, and admin setup, see the dedicated Amazon Q Business guide.
| Dimension | Q Developer | Q Business |
|---|---|---|
| Primary job | Write, fix, test & upgrade code | Answer questions over your company data |
| Buyer | Engineering / platform teams | CIO / IT / knowledge management |
| Surfaces | IDE, CLI, AWS console | Web app, Slack/Teams, embedded widget |
| Data source | Your codebase + AWS account | ~40+ enterprise connectors (RAG) |
| Pricing tiers | Free / Pro ($19/user/mo) | Lite ($3) / Pro ($20) + index cost |
| Closest competitor | GitHub Copilot, Cursor | Microsoft 365 Copilot, Glean, ChatGPT Enterprise |
| Trains base model on your data? | No | No |
Both products are thin, opinionated layers over Amazon Bedrock plus AWS's identity and data plumbing. Understanding the architecture explains both the security story and the pricing model.
Foundation models via Bedrock. Amazon Q does not ship its own model that you tune. It calls foundation models hosted on Amazon Bedrock and manages model selection for you. Because the inference runs inside AWS's managed environment, prompts and responses are not used to train the underlying base models, and data stays within AWS's security boundary. This is the same guarantee that makes Bedrock attractive to regulated buyers, inherited by Q.
Identity through IAM Identity Center. Q Business (and Q Developer Pro at the organization level) authenticate users through AWS IAM Identity Center, which can federate to your existing identity provider — Okta, Microsoft Entra ID (formerly Azure AD), Ping, and others. This is what lets Q Business enforce per-user document permissions: it knows who the asker is and only retrieves content that identity is entitled to see.
Retrieval-augmented generation (RAG) for Q Business. Q Business ingests your documents into a managed index (essentially a vector + keyword store). At query time it retrieves the most relevant passages, hands them to the model as context, and the model answers using that grounding — citing the documents it drew from. This is why Q Business can answer questions about your policies and data without that data ever being baked into a model. If you want to build comparable retrieval on raw infrastructure instead of buying the packaged product, that is the domain of Bedrock Knowledge Bases and RAG on AWS.
Agents and tools. Both products go beyond chat. Q Developer's agents can plan and execute multi-step coding tasks (implement a feature, transform a codebase). Q Business can take actions in connected systems (e.g., create a Jira ticket) through plugins. These agentic capabilities are what separate Q from a simple chatbot wrapper.
Because Q is built on Bedrock and IAM Identity Center, security and identity teams can reason about it using controls they already own — VPC boundaries, IAM, CloudTrail logging, data-residency by region. That “it’s already inside our AWS account” property is frequently the deciding factor against an external SaaS assistant.
Beyond the two licensed products, "Q" appears as an embedded capability across the AWS portfolio. Here is the full footprint so nothing surprises you.
The pattern: Q Developer is the assistant wherever engineers work (IDE, CLI, console, dev-focused chat), while Q Business is the assistant wherever knowledge workers ask questions of company data (web app, intranet widget, Slack/Teams). The embedded flavors inside QuickSight and Connect are bundled into those services' own pricing rather than the standalone Q SKUs.
Both products are per-seat with a free entry point. The figures below are representative as of 2026 — always confirm current rates on the AWS pricing page, since AWS adjusts tiers periodically.
Amazon Q Developer uses a simple two-tier model. The Free tier gives individual developers inline code suggestions, chat, and a capped number of agent interactions and security scans per month at no cost. The Pro tier is roughly $19 per user per month and adds higher limits, organization-wide license management through IAM Identity Center, policy controls, and higher caps on the agentic features (/dev, /transform) and security scanning. Some advanced agent actions (for example large-scale code transformations) can carry additional usage-based charges beyond the seat fee.
Amazon Q Business has two seat tiers plus an index cost. Q Business Lite is about $3 per user per month and is intended for lighter, mostly read/ask usage. Q Business Pro is about $20 per user per month and unlocks the full feature set including Q Apps and richer plugin actions. Separately, you pay for the index that stores and serves your data — billed by capacity units sized to how many documents and how much query throughput you need. So a Q Business bill = (seats × tier price) + (index capacity).
The index cost is the line item teams most often forget when modeling Q Business. For a small deployment it is modest; for an enterprise indexing millions of documents it becomes a meaningful share of the total. Size the index to real document volume rather than over-provisioning on day one.
| Product | Tier | Approx. price | What you get |
|---|---|---|---|
| Q Developer | Free | $0 | Inline completion, chat, capped agent runs & security scans |
| Q Developer | Pro | ~$19 / user / mo | Higher limits, org license mgmt, policy controls, more agent + scan capacity |
| Q Business | Lite | ~$3 / user / mo | Ask/answer over connected data, lighter usage |
| Q Business | Pro | ~$20 / user / mo | Full feature set incl. Q Apps & richer plugin actions |
| Q Business | Index | usage-based | Storage + query capacity for your indexed documents (billed separately) |
Because Q spans two product categories, it competes in two different markets at once. Here is the high-level read; the dedicated comparison page goes deeper on the coding head-to-head.
For coding — Q Developer vs GitHub Copilot (and Cursor). GitHub Copilot is the market-leading AI pair-programmer, deeply tied to GitHub and now multi-model. Q Developer's differentiators are its AWS awareness (it reasons about your AWS account and architecture), its agentic upgrades via /transform (e.g. Java version migrations), and its presence in the AWS console and CLI. If your stack is AWS-heavy and you value managed code transformations, Q Developer is compelling; if your team lives in GitHub and wants the broadest IDE ecosystem and model choice, Copilot is the default. The detailed breakdown lives in Amazon Q vs GitHub Copilot.
For enterprise knowledge — Q Business vs Microsoft 365 Copilot, ChatGPT Enterprise & Glean. Microsoft 365 Copilot is the natural choice for organizations standardized on Microsoft 365 (it is grounded in the Microsoft Graph — your Outlook, Teams, SharePoint). ChatGPT Enterprise is a general-purpose assistant with optional connectors and strong raw model quality. Glean is a dedicated enterprise-search-and-assistant specialist. Q Business's edge is for organizations whose data and security perimeter already live in AWS, and who want answers grounded in a broad set of connectors with per-user permission enforcement, billed and governed inside their AWS account.
The honest summary: Amazon Q is rarely the obvious pick on raw model quality alone. It wins on integration and governance — when "it runs inside our AWS account, under our IAM, in our region, without our data training someone’s model" is the requirement that matters most.
For most buyers this section is the decision. Q inherits Bedrock's data posture and AWS's identity and governance controls.
The net effect for a security review: Amazon Q can largely be evaluated with the same controls and assurances a team already applies to the rest of its AWS estate. That continuity — rather than onboarding a brand-new external processor — is a recurring reason regulated and enterprise buyers shortlist Q.
The fastest path depends on which product you need. Both start free or small and scale per seat.
For a production Q Business rollout — connector configuration across many sources, permission mapping, index sizing, guardrails, and change management — this is the kind of engagement CloudRoute routes to a vetted AWS partner, with AWS credits covering the build. See the next section for how that works.
Install the Amazon Q extension in your IDE (VS Code, JetBrains, Visual Studio, or Eclipse) or the CLI. Sign in with a free Builder ID to use the Free tier immediately — no AWS account required to try it. To roll it out across a team on the Pro tier, set up subscriptions through IAM Identity Center so licenses, policies, and limits are managed centrally.
In the AWS console, create a Q Business application, connect it to your identity provider via IAM Identity Center, then add data-source connectors (start with one or two high-value sources such as SharePoint or Confluence). Configure the index capacity, set guardrails and access controls, then share the web experience with a pilot group before expanding. Because Q Business spans data, identity, and security domains, many teams scope a focused pilot first rather than connecting everything at once.
A single scannable read across both markets Amazon Q plays in. Use it to place Q against the alternative your team is also weighing.
| Product | Category | Best when | Pricing model | Data not used to train? | Standout strength |
|---|---|---|---|---|---|
| Amazon Q Developer | AI coding assistant | AWS-heavy stack; want managed code upgrades | Free / ~$19 per user/mo | Yes | AWS awareness + /transform upgrades |
| Amazon Q Business | Enterprise RAG assistant | Data & security perimeter live in AWS | ~$3 / ~$20 per user/mo + index | Yes | Per-user permissions + 40+ connectors in your account |
| GitHub Copilot | AI coding assistant | GitHub-centric teams; broad IDE + model choice | ~$10–$39 per user/mo | Yes (Business/Enterprise) | Market-leading ecosystem & adoption |
| Microsoft 365 Copilot | Enterprise assistant | Standardized on Microsoft 365 | ~$30 per user/mo | Yes | Native Microsoft Graph grounding |
| ChatGPT Enterprise | General assistant | Want top raw model quality + flexibility | Custom / per-seat | Yes | Frontier model quality, broad tooling |
Situation: Internal support was drowning: agents and underwriters asked the same policy and procedure questions repeatedly, and answers were scattered across SharePoint, Confluence, and an S3 archive of PDF policy documents. Microsoft 365 Copilot covered the SharePoint content but not the S3 archive or the Salesforce records, and a separate external assistant was a non-starter for the security team because the policy data could not leave the company’s AWS perimeter. They wanted one assistant, grounded in all their sources, with strict per-user permissions — and they had no internal team free to build it.
What CloudRoute did: Routed within a day to an AWS Advanced-tier partner with IAM Identity Center and Q Business experience. CloudRoute helped the partner file for AWS credits to fund the engagement: a Bedrock/GenAI proof-of-concept pool to cover the pilot, with Activate Portfolio credits behind it for the broader AWS spend. The partner stood up a Q Business application, federated identity to the carrier’s existing Entra ID, connected SharePoint, Confluence, S3, and Salesforce, mapped permissions so each role only saw entitled documents, and configured guardrails before a 60-person pilot.
Outcome: Pilot answered ~80% of routine policy questions with citations inside four weeks; the carrier expanded to ~600 Pro seats over the following quarter. Roughly the first $40K of AWS consumption across the pilot and rollout was credit-funded. CloudRoute’s commission was paid by the partner out of AWS’s engagement funding — the customer paid $0 to CloudRoute.
pilot window: 4 weeks · seats at expansion: ~600 · credit-funded AWS spend: ~$40K · cost to customer: $0
CloudRoute routes you to a vetted AWS partner who builds Q Business — connectors, identity, permissions, guardrails — and helps secure AWS credits to fund it. Customer pays $0. AWS funds the engagement.