The short version

Last Light is not in the same category as most "AI coding agent" tools, and comparing it feature-for-feature is misleading. It's a self-hosted, event-driven, single-tenant software factory: a YAML workflow engine that reacts to GitHub webhooks, Slack messages, and cron ticks by running sandboxed agents on a production line to triage issues, review PRs, build features, and report on health and security.

Almost everything it's compared to is either a developer-driven pair programmer you sit in front of (Cursor, Claude Code, Codex CLI, Pi, OpenCode, Aider, Cline), or a vendor-hosted issue→PR agent you delegate to in someone else's cloud (Devin, Copilot coding agent, Codex cloud, Jules), or its commercial-platform siblings, Factory droid and 8090.

The important exceptions, and Last Light's truest peers, are the open-source agents that also run self-hosted and autonomously: OpenHands (Resolver / headless), Goose (scheduler / recipes), and Archon (an OSS workflow-harness builder). Last Light is not alone in that quadrant. The honest distinction from those is shape, covered on each comparison page.

Head-to-head

Each of these gets a full page: what the other tool does better, what Last Light does better, and when to pick which.

The category map

A flat feature table would flatter or unfairly penalise everything, because these tools live on two axes: who drives (you, per task ↔ it reacts to events) and who hosts (a vendor ↔ you). Last Light lives in the self-hosted, event-driven quadrant, but it shares it.

developer-driven (you drive it, per task)  ↔  event-driven / autonomous (it reacts)
Hosted · developer-driven

Cursor, Claude Code, Codex CLI, Windsurf

Hosted · event-driven

Devin, Copilot coding agent, Codex cloud, Jules, Factory droid, 8090

Self-hosted · developer-driven

Pi, OpenCode, Aider, Cline · OpenHands / Goose (interactive)

Self-hosted · event-driven

★ Last Light, Archon, OpenHands (Resolver), Goose (scheduler / recipes)

So the meaningful question isn't "is anything else self-hosted and autonomous" (things are) but what shape that autonomy takes: the vendor bots are autonomous but hosted; OpenHands / Goose are autonomous and self-hostable but as a single pipeline or a general agent you compose; Archon is a self-hosted harness that orchestrates external CLIs; Last Light ships a purpose-built, multi-workflow software factory for repo maintenance. That breadth is the point, and also why it carries operational burden a per-task CLI or a SaaS bot doesn't.

At a glance

Treat each cell as a generalisation, not a spec sheet. See each tool's own docs for specifics, and the sources for where these claims come from.

ToolCategoryHostingTrigger modelBYO model?IsolationHuman gatesCost modelBest for
Last LightSelf-hosted software factorySelf-hosted onlyWebhooks + Slack + cron + CLIYes (Anthropic / OpenAI / OpenRouter)Docker / micro-VM sandbox + default-deny egress firewallFirst-class (GitHub / Slack / dashboard)Free / OSS + your model spend + your infraSelf-hosted autonomous repo maintenance with approval gates
Factory droidCommercial multi-agent platformSaaS (on-prem at Enterprise)IDE + CLI + browser + Slack + Jira/LinearYes (Claude / GPT / Gemini / DeepSeek / Qwen)Managed cloud sandboxes (“Droid Computers”)Tiered autonomy levels$20 / $100 / $200 per mo + Team/EnterpriseEnterprises wanting a polished multi-surface droid fleet
8090Enterprise software factoryVendor-hosted or delivered as a serviceBusiness intent → work ordersVendor-managedVendor-managedGovernance + audit trail by designEnterprise contractRegulated enterprises modernising legacy applications
DevinAutonomous cloud agentSaaSTask / issue assignmentNo (vendor model)Vendor cloud sandboxReview the PRSubscriptionDelegating well-scoped tasks to a “junior engineer”
Copilot coding agentAutonomous GitHub agentSaaS (GitHub)Assign a GitHub issueNo (vendor model)Sandboxed GitHub ActionsReview the draft PRCopilot subscriptionTeams already living in GitHub
OpenAI Codex (cloud)Autonomous cloud agentSaaSTask / multi-surfaceNo (vendor model)Vendor cloud sandboxReview the PRSubscriptionGPT-5.x-grade autonomy
Google JulesAutonomous cloud agentSaaSGitHub issue→PRNo (Gemini)Google Cloud VMsReview the PRFree preview / subscriptionGemini-grade autonomy in GCP
OpenHandsOSS agent + ResolverSelf-hosted or cloudCLI/IDE and event-driven (issue label → PR)Yes (full BYOM)Sandboxed runtimeIn the loop or review the PRFree / OSS + your model spend (or Cloud)Self-hosted single-pipeline issue→PR autonomy
GooseOSS general agentSelf-hostedCLI/IDE, recipes, built-in schedulerYes (full BYOM)Local / your choiceConfigurable approval gatesFree / OSS + your model spendComposing your own autonomous / scheduled agent
ArchonOSS harness builderSelf-hostedTerminal + Slack + Telegram + GitHub comments + webYes (orchestrates Claude Code / Codex / etc.)Per-run Git worktreesWorkflow gatesFree / OSS + your model spendDispatching parallel coding runs from anywhere
Pi / OpenCode / Aider / ClineOSS dev-driven agentsSelf-hostableCLI / IDE, per taskYes (full BYOM)Local / your choiceYou’re in the loopFree / OSS + your model spendFree, private, hackable per-task coding

Honest limitations of Last Light

The flip side of the design choices above, stated plainly.

Decision guide

Write code interactively at your deskCursor / Claude Code / Codex CLI (or OSS: Pi / OpenCode / Aider / Cline / Goose)
Delegate a task and get a PR back, zero ops, best modelsDevin / Copilot coding agent / Codex cloud / Jules
A polished, supported, multi-surface platform for a whole teamFactory droid
An enterprise programme that builds and runs the application for you, under contract8090
A self-hosted, single-pipeline issue→PR agentOpenHands (Resolver)
A lightweight, agent-agnostic harness to dispatch parallel runs from many channelsArchon
Autonomous repo maintenance on your own infra: your runtime and keys, a boundary you control, forkable YAML, a triage/review/health/security suite, and approval gatesLast Light
A free, private, per-task agent on your own machinePi / OpenCode / Aider / Cline / Goose / OpenHands

Sources

External claims about other tools are point-in-time (June 2026). Last Light's own specifics are drawn from the spec and the project repository.

Pricing, model menus, and benchmark scores change frequently, so verify against each vendor's own pages before relying on a specific figure.