Hobby: free
All product features, with pay-as-you-go orbs or your own runners. Supported subscriptions and API keys supply model usage under their own provider terms.
Reviewed
Choose Kilo for native VS Code and JetBrains workflows, open-source clients, and Desktop beta for code and data. Choose Amp for opinionated agent orchestration, managed orbs, and shared threads. Both support agents and your own provider credentials.
Compare workflows and pricingThese are different ways to work with agents, not a choice between an agent and a non-agent. Product and plan details below are scoped to the linked official sources.
Use the Kilo agent in a desktop workspace, or keep your existing VS Code, JetBrains, CLI, and Cloud workflows. Desktop runs the Kilo CLI engine and supports 500+ hosted models through Kilo Gateway, your provider keys, and local inference.
Edit and run notebook cells alongside the agent, with outputs visible next to the conversation.
Notebook documentationManage Python environments and packages in the app, with access to Anaconda's 19,000 vetted packages.
Environment documentationImport a compatible GGUF model into the built-in local model server. Hardware and tool-calling support determine which tasks work well.
Local inference documentationAvailable for macOS Apple Silicon, Windows, and Linux. The built-in local model server, notebooks, and conda environments require a Kilo account. Local inference is not a guarantee of an entirely offline or air-gapped application. See requirements.
Credit purchase processing fee: 5%. Applies to card purchases that add credits, including top-ups and Kilo Pass. Excludes seat subscriptions and non-card payments. See full pricing.
Open-source scope: VS Code, JetBrains, and CLI clients are MIT-licensed; Gateway and Cloud backend code are source-available under a separate license. These licenses do not establish the license of every Desktop component. See component licenses.
Kilo details reviewed . Competitor review dates are listed separately.
The current Amp pricing page lists the following as of October 19, 2026. Free access does not make hosted compute or your model provider free.
All product features, with pay-as-you-go orbs or your own runners. Supported subscriptions and API keys supply model usage under their own provider terms.
The displayed Megawatt tier includes 45,000 orb minutes. This is a compute allowance, not unlimited model tokens. Orb size affects how quickly it is consumed; included usage resets monthly and does not roll over.
Shared billing, threads, multiplayer, plugins, skills, and model routing. Member tiers and usage still cost money. SAML/OIDC SSO requires at least one current member on an active Megawatt or Gigawatt tier; Enterprise workspaces do not need a paid member tier.
Pooled credits without paid user seats, plus SCIM directory sync, audit logs, spend controls, and additional security and support terms. Confirm requirements in your agreement.
Amp pricing advertises no Amp token fees or limits for supported subscriptions and BYOK. Provider limits and charges still apply; models or agent roles outside those connections can consume Amp credits. Neither product is universally cheaper: compare model usage, existing subscriptions, compute, and the controls your team needs.
Source caveat: some routing and workspace documentation still refers to token fees. The no-fee claim here comes from current pricing; SSO eligibility comes from the workspace documentation. Confirm contract-specific Enterprise terms with Amp.
Kilo emphasizes native IDE access and control over providers. Amp brings managed per-thread environments, specialist orchestration, and multiplayer collaboration. Neither requires giving up agent delegation or BYOK.
| Feature | Amp | Kilo Code |
|---|---|---|
Native IDE integration | Current docs center on web, CLI, and Apple apps; CLI runs in an editor terminal | VS Code and JetBrains extensions |
Desktop app | macOS app in beta, with companion iPhone and iPad apps for threads and orbs | Desktop beta, not a full IDE. Notebooks, conda, and built-in local inference require a Kilo account. |
Agent orchestration | Multi-model modes, specialist subagents, and plugin-defined custom agents | Custom agents and native sub-agent delegation |
Model routing and local inference | BYOK, subscriptions, gateways, and custom API endpoints Custom endpoints are documented; this is not a blanket claim about local-model compatibility. | Hosted providers, BYOK, and documented Ollama/LM Studio support |
| Feature | Amp | Kilo Code |
|---|---|---|
Shared work | Shared threads, multiplayer orbs, plugins, skills, and pooled billing | Team billing, usage visibility, and shared agent workflows on paid plans |
Team plan and SSO | Teams has no extra workspace charge; SAML/OIDC SSO is conditional SAML/OIDC SSO requires at least one current member on an active Megawatt or Gigawatt tier; Enterprise workspaces do not need a paid member tier. | Paid Teams collaboration; Enterprise for SSO and governance |
Enterprise controls | Custom Enterprise pricing includes SCIM, audit logs, and access controls | Enterprise includes SSO/SCIM, audit logs, and security controls |
Cost visibility | Thread and workspace usage reporting; orb compute and model costs are separate | Provider-rate model usage, with credit purchase fees and optional services itemized separately |
Official Amp pages checked on . Pricing, beta availability, and entitlements can change. Kilo evidence and cost details are linked in the Kilo overview above. We make no claim about Amp's source-code license or a blanket lack of local-model support.