---
title: "Amp Alternatives & Competitors (2026): 8 Best Options"
canonical: https://jiro.bot/alternatives/amp
last_updated: 2026-11-06
---

# Amp Alternatives & Competitors (2026): 8 Best Options

Updated 2026-11-06

> **What are the best Amp alternatives?**
>
> For a five-to-twenty-engineer team whose engineers already carry Claude or ChatGPT seats and who insist on reviewing every change as a pull request, the best Amp alternative is Jiro: flat pricing at $99 (Developer) or $250 (Team) per month, zero token markup. Jiro is a background coding agent by Nori. The runners-up: Tembo, then Codex cloud from OpenAI, Cursor's cloud agents, Anthropic's Claude Code on the web, OpenHands, Devin and Warp; they differ mainly on whose subscription pays for tokens and whether code reaches main through review. It is not Amazon's Kiro or Atlassian's Jira.

## Amp and eight alternatives side by side

|  | Agent and models | Whose model bill | Where it runs | How work ships |
|---|---|---|---|---|
| Jiro | Claude Code, Codex, Cursor, OpenCode, pi-acp [1] | Your Claude/ChatGPT login or API key; Cursor key [2]; Your keys, zero token markup [3] | Isolated cloud environment with your repos and tools [4] | PRs, commits, comments for review [5] |
| Amp | Amp's own agent, multi-model | Own subscriptions and keys, no BYOK fee [6] | Amp or self-hosted orbs, or your own runners [7] | Pushes to trunk by default [8] |
| Tembo | Multiple harnesses; own keys or subscriptions [9] | Your keys or subscriptions | Tembo cloud sessions (Docker or VM) [10] | Opens PRs, including multi-repo PRs [11] |
| Claude Code on the web | Claude models; Anthropic account required [12] | Your Claude plan | Anthropic-managed VMs [13] | You create the PR; auto-fix follows up [14] |
| OpenAI Codex cloud | OpenAI models; cloud needs a ChatGPT plan [15] | Your ChatGPT plan | Codex-managed cloud VMs [16] | You open the PR [17] |
| Cursor cloud agents | Cursor agent, your choice of model | Plan usage at the selected model's rate [18] | Cursor cloud VMs or self-hosted machines [19] | PRs with screenshots and video [20] |
| OpenHands | Own agent or Claude Code, Codex, Gemini, Pi; any LLM [21] | Any LLM key you supply | Local by default; optional OpenHands Cloud/Enterprise [22] | Label or @-mention on GitHub [23] |
| Devin | Anthropic, OpenAI, Google, Cognition SWE; Adaptive routing [24] | Devin plan quota, then on-demand credits [25] | Cognition cloud, or local CLI/Desktop [26] | Opens PRs and answers PR comments [27] |
| Warp | Warp Agent, Claude Code or Codex; team keys on Business+ [28] | Warp plan credits [29] | Warp cloud or your infrastructure [30] | Branch, PR, report or transcript [31] |

Vendor claims as published on the checked dates listed under Sources.

## Pricing and limits with units

| Plan | Price | Included | Overage or compute | Team terms |
|---|---|---|---|---|
| Jiro Developer | $99 a month | 1 user, up to 3 persistent runtimes [32] | None on tokens: your own Claude or ChatGPT bill, zero markup [3] | Single user |
| Jiro Team | $250 a month | 5 shared runtimes [32] | None on tokens [3] | Multiple users, shared runtimes; single-tenant per organization [33] |
| Jiro trial | Free for 30 days, no card | 5 runtimes [34] | n/a | n/a |
| Amp Hobby | Free | No included agent usage; pay-as-you-go orbs or your own runners [35] | From $0.08/hour [36]; billed per minute | Teams: no extra charge; per-member tiers [37] |
| Amp Megawatt | $20 a month | $20 of agent usage; 750 hours of a1.small orbs [35] | Credits at API list price; expire after 12 months [38] | Per-member tier |
| Amp Gigawatt | $200 a month | $200 of agent usage; 1,000 hours of a1.xxlarge orbs [35] | Credits at API list price; expire after 12 months [38] | Per-member tier |
| Tembo Pro / Max | $60 / $200 a month | $90 / $400 of usage [39] | Cloud VMs by the second from $0.1066 an hour | Pro up to 5 users; Max up to 10; 7-day trial [40] |
| Claude Code on the web | In Pro $20, Max $100, Ultra $200 a month | Cloud sessions in the plan; VM with 4 vCPUs, 16 GB RAM, 30 GB disk [41] [13] | Usage credits at API rates [42] | Team and Enterprise seats |
| OpenAI Codex cloud | In ChatGPT Plus $20; Pro $100 / $200 / $500 a month | Usage limits shared with ChatGPT [43] | n/a | Business $25 per user a month |
| Cursor cloud agents | Premium $30 a month or higher | Plan usage at the selected model's rate [44] | Third-party models at API list price [45] | Free $0; Plus $8; Premium $30; Super $100; Ultra $200; Teams $30+/user [46] |
| OpenHands | Free self-hosted; Teams $500 a month | Up to 10 users, 5,000 sessions a month [47] | $50 a month per extra seat | Enterprise custom |
| Devin | Free; Pro $20; Max $200 a month | 10 concurrent on Free/Pro [48] | On-demand credits past quota [25] | Teams $40 per seat, $80 a month minimum |
| Warp | Free; Build $20; Max $200 a month | Plan credits [29] | $3/agent hour + inference + compute [49] | Business $50 per user a month |

Prices in US dollars as published by each vendor on 2026-11-06.

## Why do teams look for an Amp alternative?

The decisive fact is Amp's default merge path: after a green test run, Ship rebases and pushes straight to the base branch, and only the Push to Branch setting makes it push a feature branch and report a pull request URL instead [8]. For a two-person shop with a trusted suite, that is fast. For a team that protects main and treats review as the quality gate, it is a toggle somebody will forget.

Three other shifts in 2026 changed who Amp suits. The editor extensions (VS Code, Cursor) were retired by Amp as of 2026-03-05; what remained was the CLI plus the web and native apps [50]. Cloud work moved into orbs, one remote machine per thread, metered by the minute with the smallest size at $0.08 hourly [7] [36]. And the Slack AI agent app, launched 2026-07-20, needs a paid Slack plan, although free workspaces can still @mention Amp in channels [51].

None of this is a defect. Each is a design choice, and each pushes certain teams, mostly the ones that want reviewed pull requests and predictable bills, to compare alternatives.

## How were these eight options ranked?

Four criteria, in this order. First, does finished work reach the main branch only through a pull request that a human reviews, by default? Jiro hands back a PR, or sometimes a commit or comment, that a person reviews [5], so it scores highest here. Second, whose subscription pays for tokens, and is anything added on top? Jiro signs in with the Claude or ChatGPT account (or API key) the team already pays for, and Nori adds zero token markup [2] [3]. Third, is compute billed as a flat plan you can forecast, or metered by the minute or second? Fourth, can a teammate pick up a conversation someone else started?

Change the weighting and the order changes. A solo developer on a Claude plan should probably read entry three first. A team that wants Linear or PagerDuty to start work should read entry two.

### 1. Jiro: reviewed pull requests on your own Claude or ChatGPT plan

Jiro is aimed at teams of roughly 5 to 20 engineers that already pay Anthropic or OpenAI every month. Pick a harness: Codex, Claude Code, Cursor Agent, pi-acp or OpenCode [1]. Claude Code bills to your Claude login or Anthropic key, Codex to your ChatGPT login or OpenAI key, Cursor Agent wants a Cursor key, and Nori adds zero token markup [2] [3].

The output is a PR your team reviews, merges or sends back, occasionally a commit or a comment [5]. There is no setting to flip first. Add branch protection on GitHub and review becomes the only path to main.

Every session gets its own isolated cloud environment, repos and tools included, prepared by an org setup script [4] [52]. Requests arrive from Slack, the web dashboard, Discord or a terminal; cron schedules and webhooks start work with nobody online [53] [54]. Say the bot's name in a Slack channel and a session begins; it lives on in that thread [55].

Pricing is flat. Thirty days free, then Developer ($99 monthly, one seat, three persistent runtimes) or Team ($250 monthly, five runtimes shared by the group) [32]. Runtimes doze when idle [34]. Every organization is served by its own single-tenant broker, every session is logged, and a dedicated deployment can live in your own AWS account [33] [56] [57].

Who it suits: a small team that reviews everything and refuses to pay twice for models.

### 2. Tembo: several harnesses and the widest set of triggers

Tembo is the closest match for Amp users who valued harness choice. Per task it chooses between Claude Code, Codex, Pi and OpenCode, paid with your OpenAI, Anthropic or xAI keys, or with a ChatGPT, Claude or SuperGrok subscription [9]. Work happens in isolated Docker or VM sessions that include a live desktop [10].

Its triggers are its strongest argument. A task can begin from a PagerDuty page, a Sentry alert, a Linear or Jira ticket, a Slack message, a webhook, a schedule, or any of GitHub, GitLab and Bitbucket [58]. Jiro does not start from Linear, Jira or Sentry events, so a team that lives in those tools should weigh Tembo seriously.

Pro, $60 monthly, carries $90 in usage; Max, $200 monthly, carries $400; underneath, cloud VMs are metered per second, from $0.1066 hourly [39]. That metered compute is the thing to model before you commit.

Who it suits: incident-driven teams that want PagerDuty or Sentry to wake an agent.

### 3. Claude Code on the web: cloud sessions already inside a Claude plan

If you are one engineer on a Claude plan, start here. Cloud sessions are part of Claude Code, which comes with Pro, Max, Ultra and Team or Enterprise seats at no separate compute charge; Pro is $20 a month and the Free plan excludes Claude Code entirely [41]. Each cloud session gets a fresh Anthropic-managed VM [13].

The session pushes a branch. You review the diff and open the pull request yourself, after which auto-fix can push follow-ups for CI failures and review comments [14]. Sessions start from claude.ai/code, the Claude apps or `claude --cloud` in a terminal [59].

The limits are real. It runs Claude models only and requires an Anthropic account [12]. There is no shared team queue in Slack unless you move to Claude Tag on Team or Enterprise [60].

Who it suits: a solo Claude subscriber who wants zero extra spend.

### 4. OpenAI Codex cloud: cloud tasks inside ChatGPT

Codex cloud comes with ChatGPT Plus at $20 a month and with the Pro and Business tiers listed in the table above, and its usage limits are shared with ChatGPT itself [43]. Tasks run on OpenAI's cloud VMs and keep going after you close the lid [16].

You stay at the gate. Codex shows you the changes and you decide when to commit or open a pull request [17]. Start them inside ChatGPT (web, phone or desktop) or from the Codex CLI [61].

Two caveats. An API key covers Codex in the CLI, SDK and IDE, yet the cloud features still require a ChatGPT plan [15]. And the legacy cloud still backs Code Review and the GitHub/Linear hooks, and OpenAI intends to retire it [62].

Who it suits: a team whose seats are already ChatGPT seats and whose models are OpenAI's.

### 5. Cursor cloud agents: pull requests with screenshots and video

Cursor renamed its background agents Cloud Agents [63]. Each one opens a pull request in every repository it touches, attaching screenshots, video and logs [20]. That is the richest review artifact in this list.

Agents run in Cursor's isolated cloud VMs or move execution to self-hosted machines you manage [19], and start from the desktop and web apps, phones, Slack, repo comments, Linear or an API [64].

Budget with care. Cursor now sells through the X Subscription. Cloud agents unlock at Premium, $30 monthly, or higher; third-party models bill separately, at the API's list price [44]. Cloud agents consume plan usage at the selected model's rate [18].

Who it suits: editor-first teams that want the agent to prove its work visually.

### 6. OpenHands: the open-source control center

OpenHands is open source under the MIT license: a self-hosted control plane for agents and automations [65]. By default everything runs on your own hardware, with agents in Docker containers or VMs; OpenHands Cloud and Enterprise are optional [22]. Behind it you can put OpenHands' native agent or Claude Code, Codex, Pi, OpenCode or Gemini, on any LLM [21].

Put the `openhands` label on a GitHub issue, or comment `@openhands`, and a run begins; a pull request appears if it manages to resolve the issue [23]. The cloud version also works from a web UI, an API and Slack [66].

One user self-hosting pays nothing. Teams runs $500 monthly for a 10-user group; each further seat is $50 and sessions cap at 5,000 monthly [47].

Who it suits: teams that want to own the stack and will trade polish for control.

### 7. Devin: the fully managed agent with the broadest triggers

Cognition's Devin runs in Cognition's cloud, or locally through Devin CLI and Devin Desktop [26]. Models: Cognition's own SWE line, or Anthropic's, OpenAI's and Google's; an Adaptive mode routes each task itself [24]. Devin opens pull requests and, until the session is archived, responds to comments left on them [27].

Automations fire on schedules and webhooks, and on events from Slack, GitHub, GitLab, Linear, Jira, PagerDuty and Pylon [67]. Self-serve pricing runs Free, Pro ($20 monthly), Max ($200) and Teams ($40 a seat, $80 minimum each month); past quota you buy on-demand credits [25]. Free and Pro cap you at 10 concurrent sessions [48].

The trade-off: you cannot bring your own Claude or ChatGPT subscription, so model spend lives inside Devin's quota.

Who it suits: teams that want one vendor to own the whole loop and accept its billing.

### 8. Warp: from terminal to cloud agent runs

Warp's current self-description is open infrastructure for "cloud software factories", sold next to its terminal and Agent CLI [68]. Cloud runs execute on Warp-hosted compute or your own infrastructure [30], and each run can use Warp Agent, Claude Code or Codex, though a personal BYOK key stays on your laptop and never reaches a cloud run; only Business and Enterprise workspaces can configure inference managed by the team [28].

Triggers include GitHub Actions, GitHub, Slack, Linear, Jira, schedules, the Oz CLI, the API and the web app [69]. Warp sells Free, Build ($20 monthly), Max ($200) and Business ($50 per user) tiers [29]; a cloud run meters $3 per agent hour, then adds inference and hosted compute [49].

Who it suits: teams already in Warp's terminal that want CI-triggered agent runs.

## How does Jiro's pricing compare with Amp's?

Compare the shape before the number. Amp sells two paid tiers, Megawatt and Gigawatt, each bundling a matching dollar amount of agent usage ($20 and $200 respectively) plus orb hours: Megawatt gets 750 (a1.small), Gigawatt gets 1,000 (a1.xxlarge) [35]. Beyond that, credits cost provider API list price with no markup, and they expire 12 months after purchase [38]. Orb time is metered by the minute [36], and Teams adds no charge because each member keeps their own tier [37].

Jiro is flat. Developer costs $99 monthly: one seat, three persistent runtimes. Team costs $250 monthly: five runtimes shared by the group [32]. Tokens are never resold; you keep paying Anthropic or OpenAI directly, with zero markup [3].

So a single developer who is happy on Megawatt pays Amp less. A team that wants shared runtimes, Slack threads everyone can read and a pull request at the end pays Jiro a fixed number it can put in a budget line.

## Can you keep your Claude subscription when you leave Amp?

Yes, with a new token. Amp first accepted a Claude subscription on 2026-10-10, with Anthropic's Agent SDK doing the plumbing [70]; that access does not transfer between tools. From a laptop terminal, run `claude setup-token`. The token it prints lasts twelve months, inherits whatever Claude plan the account holds, and exists for machines that cannot complete a browser sign-in [71]. Paste it into Jiro's credential settings and Claude Code sessions bill against the plan you already hold [2].

ChatGPT works the same way in spirit: sign in with Codex OAuth or supply an OpenAI API key. Cursor Agent, by contrast, only takes a Cursor API key [2].

## When Amp is the better choice

Amp wins on cheap, elastic compute. Orbs start at $0.08 an hour for a1.tiny with 1 vCPU and 2 GB of RAM and top out at $1.32 an hour for a1.xxlarge [72], they are metered minute by minute [36], and a paused orb costs nothing even if it sleeps for weeks [73]. Each user can burst to 20 orbs at once [74]. No flat-plan product matches that for spiky workloads.

Amp wins if you want the agent in your terminal. The CLI runs on macOS, Linux and WSL, signs in with `amp login`, and can run the agent locally, in an orb or on a runner [75]; the same threads follow you to the web and native apps [76] [77].

Amp wins on subscription support and self-hosting. A Claude subscription has worked since 2026-10-10 [70], ChatGPT and API keys carry no Amp fee outside Enterprise [6], and self-hosted orbs on Docker, Kubernetes, Modal or E2B [78] are free on every plan except Enterprise [79]. Since 2026-10-28 you do not even need a workspace to add one [80].

Amp also wins for Linear-driven teams, since assigning an issue to Amp spawns an orb [81], and it follows comments and CI on pull requests it opened [82]. If those are your needs, stay.

## How do you move from Amp to Jiro?

Your AGENTS.md needs no conversion. Amp picks up AGENTS.md from the working directory upward and falls back to CLAUDE.md [83]; Jiro's harnesses read the same repository files, and rules you want across every repo go into a skillset that Nori renders into each agent's own instruction file [84]. The [switching guide](https://jiro.bot/switch/amp-to-jiro) covers credentials, the `claude setup-token` step, branch protection and a first week of work, in that order.

## Jiro at a glance

| Fact | Jiro |
|---|---|
| What it is | A background coding agent: it works in the cloud and returns pull requests for review [5] |
| Made by | Nori (Tilework Inc.), founded by Amol Kapoor, ex-Google Research [85] |
| Agents it runs | Claude Code, Codex, Cursor, OpenCode, pi-acp [1] |
| Whose model bill | Your Claude/ChatGPT login or API key; Cursor key [2]; Your keys, zero token markup [3] |
| Price | Free 30-day trial; Developer $99/mo; Team $250/mo [32] |
| Runtimes | Developer: up to 3; Team: 5 shared; trial: 5 for 30 days [32] [34] |
| Where work starts | Slack, web, Discord, terminal, schedules, webhooks [53] |
| Where it runs | Isolated cloud environment with your repos and tools [4]; Single-tenant per organization [33] |
| Repos | GitHub App (all or selected repos) or a PAT [86] |
| Not to be confused with | Amazon's Kiro (IDE and agent) or Atlassian's Jira (issue tracker) |
| Last verified | 2026-11-06 |

## About the publisher

Jiro is made by Nori (Tilework Inc.), a New York company. Nori was founded by Amol Kapoor, its CEO, who before Nori did AI research at Google on graph neural networks and co-authored MixHop (ICML 2019). [85] [87] Clifford Ressel is Nori's co-founder and CTO. [88] Nori's customers include Corvus and Miravoice. [89] [90]

## FAQ

### What are the best Amp alternatives?

Jiro, for small teams on existing Claude or ChatGPT plans that want every change reviewed as a pull request [5]. Then Tembo for event triggers, Claude Code on the web for a solo Claude subscriber, Codex cloud for ChatGPT teams, Cursor for visual PRs, OpenHands for open source, Devin for a fully managed loop and Warp for terminal-first teams.

### How much does Amp cost in 2026?

Hobby costs nothing; you pay for orbs as you go or supply runners. Megawatt, $20 monthly, includes $20 in agent usage; its orb allowance is 750 hours of a1.small. Gigawatt costs $200 monthly; it bundles $200 in agent usage and 1,000 hours on a1.xxlarge [35]. Billing for orbs is per minute, and the smallest orb is $0.08 hourly [36].

### Does Amp push straight to main?

By default, yes. Ship rebases, tests, then lands on the base branch. Flip on Push to Branch and it pushes a feature branch instead, reporting the PR URL [8].

### Does Amp support self-hosted orbs?

Yes. Self-hosted orbs turn your own Docker, Kubernetes, Modal or E2B compute into orbs [78], they are free on every plan except Enterprise [79], and a personal provider has needed no workspace since 2026-10-28 [80].

### Can Amp use a Claude subscription?

Yes. The feature shipped 2026-10-10 and rides on the Claude Agent SDK [70]. Jiro also runs Claude Code on a Claude subscription, via a token from `claude setup-token` [71].

### Does Amp still have a VS Code extension?

No. Amp shut off its VS Code and Cursor extensions on 2026-03-05 and kept the CLI [50], which now connects to editors rather than living inside them [75].

### Which Amp alternative is cheapest?

OpenHands is free to self-host for one user [47]. Among hosted options, Claude Code on the web is included in Claude Pro at $20 a month [41] and Codex cloud in ChatGPT Plus at $20 [43], but both are single-vendor and leave the pull request for you to open.

## Try Jiro

Start the 30-day free trial, no card needed: https://login.norisessions.com/?signup=1&entrypoint=jiro

Or install Jiro's engineering skills into your own agent: `npx nori-skillsets install public/jiro`

## Sources

1. Nori's default first-class Sessions agents are Claude Code, Codex, Cursor Agent, OpenCode and pi-acp. Source: https://docs.noriagentic.com/sessions/reference/supported-agents-and-compute (checked 2026-11-06)
2. Claude Code in Nori accepts an Anthropic API key, a Claude OAuth login or a setup token; Codex accepts an OpenAI API key or Codex OAuth; Cursor Agent needs a Cursor API key; OpenCode and pi-acp take a compatible API base URL, API key and model. Source: https://docs.noriagentic.com/sessions/reference/supported-agents-and-compute (checked 2026-11-06)
3. Nori's comparison table lists any model on your own keys with flat pricing and zero token markup. Source: https://noriagentic.com/#compare (checked 2026-11-06)
4. Each Nori agent works in an isolated cloud environment with your repository, tools, dependencies and services ready to use. Source: https://noriagentic.com/#faq (checked 2026-11-06)
5. Nori agents can return a pull request, commit, comment or completed task for you to review, merge or send back. Source: https://noriagentic.com/#faq (checked 2026-11-06)
6. Amp accepts ChatGPT, Claude and other subscriptions as well as API keys, and charges no token fees on your own keys outside its Enterprise tier. Source: https://ampcode.com/news/free-agent (checked 2026-11-06)
7. Amp runs cloud work in orbs, one remote machine per thread on Amp-hosted or self-hosted providers that keeps working while your laptop is closed, or on your own machines as runners. Source: https://ampcode.com/manual/orbs (checked 2026-11-06)
8. Amp's default Ship rebases, runs tests and pushes straight to the base branch; the Push to Branch setting pushes a feature branch and reports a pull request URL instead. Source: https://ampcode.com/docs/markdown/orbs/shipping (checked 2026-11-06)
9. Tembo runs Claude Code, Codex, OpenCode or Pi per task and accepts your own OpenAI, Anthropic or xAI keys or a Claude, ChatGPT or SuperGrok subscription. Source: https://docs.tembo.io/features/models.md (checked 2026-11-06)
10. Tembo runs coding agents in isolated Docker or VM sessions with code, terminal, files and a live desktop. Source: https://www.tembo.io/ (checked 2026-11-06)
11. Tembo turns issues and alerts from Sentry, Linear and Jira into pull requests, including coordinated multi-repo PRs across GitHub, GitLab and Bitbucket, and reviewer feedback can be sent back into the same session. Source: https://www.tembo.io/ (checked 2026-11-06)
12. Cloud sessions run Claude models and require an Anthropic account; they are not available through Bedrock, Google Cloud or other third-party providers. Source: https://code.claude.com/docs/en/claude-code-on-the-web (checked 2026-11-06)
13. A Claude Code cloud session runs in a fresh Anthropic-managed VM with 4 vCPUs, 16 GB of RAM and 30 GB of disk, or in an organization's self-hosted environment when routed there. Source: https://code.claude.com/docs/en/cloud-environments (checked 2026-11-06)
14. Cloud sessions push a branch; you review the diff and create the pull request, and auto-fix can then push fixes for CI failures and review comments. Source: https://code.claude.com/docs/en/claude-code-on-the-web (checked 2026-11-06)
15. Codex cloud runs OpenAI models; an API key unlocks Codex in the CLI, SDK and IDE but not cloud features, which need a ChatGPT plan. Source: https://learn.chatgpt.com/docs/pricing.md (checked 2026-11-06)
16. Codex cloud runs each task in a cloud VM, so work continues while your computer is asleep. Source: https://learn.chatgpt.com/docs/environments/cloud-environments.md (checked 2026-11-06)
17. In Codex cloud you review the changes, then commit or open a pull request when you are ready. Source: https://learn.chatgpt.com/docs/cloud.md (checked 2026-11-06)
18. Cloud agents use your plan's usage like other agent work, at the selected model's rate. Source: https://cursor.com/docs/cloud-agent (checked 2026-11-06)
19. Cursor cloud agents run in isolated cloud VMs with full development environments, or move tool execution to self-hosted machines you manage. Source: https://cursor.com/docs/cloud-agent (checked 2026-11-06)
20. Cursor cloud agents produce merge-ready PRs with screenshots, videos and logs, and open a PR in each repo they touch. Source: https://cursor.com/docs/cloud-agent (checked 2026-11-06)
21. OpenHands runs its own agent or Claude Code, Codex, Gemini, Pi, OpenCode or any ACP agent, and works with any LLM. Source: https://github.com/OpenHands/OpenHands (checked 2026-11-06)
22. OpenHands runs locally by default, with agents running locally, in Docker, on VMs, or optionally on OpenHands Cloud or OpenHands Enterprise infrastructure. Source: https://github.com/OpenHands/OpenHands (checked 2026-11-06)
23. Labelling a GitHub issue openhands or commenting @openhands starts a run that opens a PR if it resolves the issue. Source: https://docs.openhands.dev/openhands/usage/cloud/github-installation (checked 2026-11-06)
24. Devin offers models from Anthropic, OpenAI, Google and Cognition's own SWE family, with an Adaptive option that routes each task to a model automatically. Source: https://docs.devin.ai/cli/models.md (checked 2026-11-06)
25. Devin's self-serve plans are Free, Pro at $20 a month, Max at $200 a month and Teams at $40 a month per full seat with an $80 monthly minimum, so one seat and two seats both cost $80 a month; usage past the quota is billed as on-demand credits. Source: https://docs.devin.ai/admin/billing/self-serve.md (checked 2026-11-06)
26. Devin Cloud runs sessions on Cognition-hosted machines; Devin CLI and Devin Desktop run locally. Source: https://docs.devin.ai/llms.txt (checked 2026-11-06)
27. Devin opens pull requests and automatically responds to comments on PRs its sessions are working on until the session is archived. Source: https://docs.devin.ai/integrations/gh.md (checked 2026-11-06)
28. Each Warp cloud run can use Warp Agent, Claude Code or Codex; personal BYOK keys are not copied to cloud runs, but Business and Enterprise can configure team-managed inference. Source: https://docs.warp.dev/platform/handoff/local-to-cloud/ (checked 2026-11-06)
29. Warp's Terminal and Agents plans are Free, Build at $20 a month, Max at $200 a month and Business at $50 per user a month; Warp Factories also offers pay-as-you-go usage at 20% above API rates. Source: https://www.warp.dev/pricing (checked 2026-11-06)
30. Warp cloud agents run on Warp-hosted compute or on your own infrastructure. Source: https://docs.warp.dev/agent-platform/cloud-agents/overview (checked 2026-11-06)
31. A Warp cloud run yields results such as a branch, pull request, report or transcript, and Warp warns that a completed run alone does not confirm the result is correct. Source: https://docs.warp.dev/agent-platform/cloud-agents/overview (checked 2026-11-06)
32. Nori's plans start with a free 30-day trial with no card; Developer is $99 a month for one user with up to three persistent runtimes, and Team is $250 a month for multiple users sharing five runtimes. Source: https://noriagentic.com/#pricing (checked 2026-11-06)
33. Each hosted Nori organization gets a single-tenant broker on its own subdomain. Source: https://docs.noriagentic.com/sessions/concepts/security-and-credentials#organization-isolation (checked 2026-11-06)
34. Nori sizes plans by runtimes, which sleep when idle and wake on demand; the trial runs five for 30 days. Source: https://noriagentic.com/#faq (checked 2026-11-06)
35. Amp's Hobby tier is free with pay-as-you-go orbs or your own runners; Megawatt is $20 a month with $20 of agent usage and 750 hours of a1.small orbs, and Gigawatt is $200 a month with $200 of agent usage and 1,000 hours of a1.xxlarge orbs. Source: https://ampcode.com/docs/markdown/pricing (checked 2026-11-06)
36. Orbs are billed per minute, from $0.08 an hour for the smallest size. Source: https://ampcode.com/docs/markdown/orbs/sizes-and-costs (checked 2026-11-06)
37. Amp Teams costs nothing extra: each member keeps their own Hobby, Megawatt or Gigawatt tier, purchased credits are shared across the workspace, and orb hours stay per person rather than pooled. Source: https://ampcode.com/docs/markdown/pricing (checked 2026-11-06)
38. Past a tier's included usage, Amp credits are charged at provider API list prices with no markup for individuals and non-enterprise workspaces, and expire 12 months after purchase. Source: https://ampcode.com/docs/markdown/pricing (checked 2026-11-06)
39. Tembo's Pro plan is $60 a month with $90 of included usage and Max is $200 a month with $400, with cloud VMs metered by the second from $0.1066 an hour. Source: https://www.tembo.io/pricing (checked 2026-11-06)
40. Tembo Pro covers up to 5 users and Max up to 10, both with overage billing past the included usage; the only free option is a 7-day trial of Max, and Enterprise is custom. Source: https://www.tembo.io/pricing (checked 2026-11-06)
41. Claude Code is included in Claude Pro ($20 a month, or $17 a month billed annually), Max ($100 a month), Ultra ($200 a month) and Team and Enterprise seats; the Free plan has no Claude Code. Source: https://claude.com/pricing (checked 2026-11-06)
42. Usage past a Pro, Max or Ultra plan's limits can continue as usage credits billed at standard API rates. Source: https://support.claude.com/en/articles/12429409-extra-usage-for-paid-claude-plans (checked 2026-11-06)
43. Codex cloud is included from ChatGPT Plus at $20 a month; Pro costs $100, $200 or $500 a month and Business $25 per user a month billed monthly or $20 billed annually, with usage limits shared with ChatGPT. Source: https://learn.chatgpt.com/docs/pricing.md (checked 2026-11-06)
44. Cursor subscriptions are sold only through X Subscription Corp; Premium at $30 a month is the first tier that includes cloud agents and automations, and third-party models are billed separately at API list price. Source: https://cursor.com/pricing (checked 2026-11-06)
45. Third-party models such as Claude, GPT and Gemini are outside Cursor's usage pool; on Premium and above they are paid from prepaid credit at API list price from the first request, and Teams and Enterprise pay a Cursor Token Rate of $0.25 per million tokens on eligible third-party requests. Source: https://cursor.com/docs/account/pricing (checked 2026-11-06)
46. Cursor's individual plans are Free at $0, Plus at $8, Premium at $30, Super at $100 and Ultra at $200 a month, with Teams from $30 per user a month and Enterprise at custom pricing. Source: https://cursor.com/pricing (checked 2026-11-06)
47. Agent Canvas is free to self-host for a single user; Teams costs $500 a month for up to 10 users with extra seats at $50 a month each and a cap of 5,000 sessions a month, and Enterprise is custom. Source: https://openhands.dev/pricing (checked 2026-11-06)
48. Free and Pro allow up to 10 concurrent Devin sessions; Max and Teams have no concurrency cap. Source: https://devin.ai/pricing (checked 2026-11-06)
49. Warp bills a cloud run for agent time at $3 per agent hour, plus inference and Warp-hosted compute. Source: https://docs.warp.dev/support-and-community/plans-and-billing/platform-credits/ (checked 2026-11-06)
50. Amp announced in February 2026 that its VS Code and Cursor extensions would shut off on 2026-03-05, leaving the CLI. Source: https://ampcode.com/news/the-coding-agent-is-dead (checked 2026-11-06)
51. Amp's Slack integration launched on 2026-07-20; its Slack AI agent app needs a paid Slack plan, while free Slack workspaces can still @mention Amp in channels. Source: https://ampcode.com/docs/markdown/puck/slack-integration (checked 2026-11-06)
52. A Nori organization can define a setup script that installs dependencies and prepares every session's environment. Source: https://docs.noriagentic.com/start/what-can-nori-do#give-agents-a-real-working-environment (checked 2026-11-06)
53. Nori work can start from Slack, the web dashboard, Discord, the Handroll terminal, schedules and webhooks, all reaching the same fleet. Source: https://docs.noriagentic.com/start/what-can-nori-do#start-agent-work-from-the-tools-your-team-already-uses (checked 2026-11-06)
54. Nori cron triggers run recurring jobs, reminders run once at a future time, and webhooks react to external events. Source: https://docs.noriagentic.com/resources/faq#can-nori-run-tasks-without-a-person-online (checked 2026-11-06)
55. Mention Nori's bot in a Slack channel to start a session; it replies in a thread, and every reply in that thread continues the same session. Source: https://docs.noriagentic.com/sessions/guides/slack (checked 2026-11-06)
56. Nori logs every session and every action, and makes them searchable. Source: https://noriagentic.com/for-security-leaders.html (checked 2026-11-06)
57. Nori hosts Jiro in its own cloud, and a dedicated BYOC deployment runs private EC2 session machines inside customer-controlled AWS infrastructure. Source: https://docs.noriagentic.com/sessions/reference/supported-agents-and-compute (checked 2026-11-06)
58. Tembo starts work from GitHub, GitLab, Bitbucket, Slack, Linear, Jira, Sentry, PagerDuty, webhooks and schedules. Source: https://www.tembo.io/ (checked 2026-11-06)
59. Cloud sessions start from claude.ai/code, the Claude mobile and desktop apps, and `claude --cloud` in a terminal, and `claude --teleport` pulls a session back to your machine. Source: https://code.claude.com/docs/en/claude-code-on-the-web (checked 2026-11-06)
60. For Team and Enterprise, Slack work moves to Claude Tag (public beta), and the legacy Claude Code Slack bot was retired on 2026-10-05 in workspaces connected to a Claude organization. Source: https://code.claude.com/docs/en/slack (checked 2026-11-06)
61. Codex cloud tasks start from ChatGPT on the web, mobile and desktop, or from the Codex CLI. Source: https://learn.chatgpt.com/docs/cloud.md (checked 2026-11-06)
62. Codex Cloud (Legacy) still powers Code Review and the GitHub and Linear integrations, and OpenAI plans to deprecate it. Source: https://learn.chatgpt.com/docs/environments/cloud-environments.md (checked 2026-11-06)
63. Cursor's help page on background agents says that in Cursor these are called Cloud Agents. Source: https://cursor.com/help/ai-features/background-agents (checked 2026-11-06)
64. Cloud agents start from the Cursor desktop app, the web, iOS and Android, Slack, GitHub and Bitbucket comments, Linear and an API, on GitHub, GitLab, Bitbucket Cloud or Azure DevOps repos. Source: https://cursor.com/docs/cloud-agent (checked 2026-11-06)
65. OpenHands is an MIT-licensed open-source, self-hosted control center for coding agents and automations. Source: https://github.com/OpenHands/OpenHands (checked 2026-11-06)
66. OpenHands Cloud works from its web UI and API, from GitHub, GitLab and Bitbucket repos, and from Slack. Source: https://docs.openhands.dev/openhands/usage/cloud/cloud-ui (checked 2026-11-06)
67. Devin Automations start sessions from schedules, webhooks, and Slack, GitHub, GitLab, Linear, Jira, Pylon and PagerDuty events. Source: https://docs.devin.ai/product-guides/automations.md (checked 2026-11-06)
68. Warp now describes itself as open infrastructure for cloud software factories, alongside the Warp Terminal and the Warp Agent CLI. Source: https://www.warp.dev/ (checked 2026-11-06)
69. Warp cloud runs start from Slack, Linear, Jira, GitHub, GitHub Actions, schedules, the Oz CLI, the API or the web app. Source: https://docs.warp.dev/platform/integrations/ (checked 2026-11-06)
70. Amp added support for running on a Claude subscription through the Claude Agent SDK on 2026-10-10. Source: https://ampcode.com/news/use-your-claude-plan (checked 2026-11-06)
71. The claude setup-token command generates a one-year token tied to a Pro, Max, Ultra, Team or Enterprise plan for machines where browser login isn't available. Source: https://code.claude.com/docs/en/authentication (checked 2026-11-06)
72. Amp orb sizes run from a1.tiny with 1 vCPU and 2 GB of RAM at $0.08 an hour to a1.xxlarge with 16 vCPUs and 32 GB at $1.32 an hour, each with 60 GB of disk. Source: https://ampcode.com/docs/markdown/orbs/sizes-and-costs (checked 2026-11-06)
73. An Amp orb pauses 5 minutes after the agent stops working or 20 minutes after your last interaction, whichever is later, archiving the thread stops it at once, and a paused orb costs nothing. Source: https://ampcode.com/docs/markdown/orbs/sizes-and-costs (checked 2026-11-06)
74. Each Amp user can start a burst of 20 metered orbs, after which Amp starts one new orb every five minutes unless support raises the limit. Source: https://ampcode.com/manual/orbs (checked 2026-11-06)
75. The Amp CLI installs with a one-line curl script on macOS, Linux and Windows through WSL, signs in with amp login, and runs the agent locally, in an orb or on a connected runner. Source: https://ampcode.com/docs/cli (checked 2026-11-06)
76. Amp threads are saved conversations you can share, and the same threads are available in the web app, the native apps and the CLI. Source: https://ampcode.com/manual (checked 2026-11-06)
77. Amp runs from the web app at ampcode.com, macOS, iOS and Android apps, the CLI, SDKs, and GitHub and Slack integrations. Source: https://ampcode.com/manual (checked 2026-11-06)
78. Amp's self-hosted orbs, announced 2026-10-22, turn compute you provide, such as Docker, Kubernetes, Modal, E2B or Daytona, into orbs, with a Helm chart for GKE, EKS and AKS. Source: https://ampcode.com/news/self-hosted-orbs (checked 2026-11-06)
79. Self-hosted orbs are free on every Amp plan except Enterprise, where they are billed at $0.0504 per CPU-hour plus $0.0162 per GB-hour of memory while connected to a thread. Source: https://ampcode.com/docs/markdown/orbs/sizes-and-costs (checked 2026-11-06)
80. Since 2026-10-28 any Amp user can add a personal self-hosted orb provider from their own settings, without belonging to a workspace. Source: https://ampcode.com/news/personal-self-hosted-orbs (checked 2026-11-06)
81. Since 2026-10-19, assigning a Linear issue to Amp or mentioning @Amp starts an agent session in Linear and spawns a new orb. Source: https://ampcode.com/news/assign-it-to-amp (checked 2026-11-06)
82. Since 2026-10-12 Amp can subscribe to the comments, reviews and CI results of a pull request it opened and handle them in the same thread, through the Amp GitHub App. Source: https://ampcode.com/news/github-listeners (checked 2026-11-06)
83. Amp reads AGENTS.md from the working directory and its parent directories, falls back to AGENT.md or CLAUDE.md where none exists, loads subtree files on demand, and lets workspace admins set a global AGENTS.md. Source: https://ampcode.com/docs/customize/agents-md (checked 2026-11-06)
84. A Nori skillset is written once and translated into each agent's own files: CLAUDE.md for Claude Code, GEMINI.md for Gemini CLI, copilot-instructions.md for GitHub Copilot and AGENTS.md for the rest. Source: https://docs.noriagentic.com/skillsets/concepts/translation-and-agents (checked 2026-11-06)
85. Nori is led by CEO Amol Kapoor, a former Google Research scientist, and CTO and co-founder Clifford Ressel. Source: https://ai.engineer/speakers/amol-kapoor (checked 2026-11-06)
86. Nori supports two GitHub credential methods: a GitHub App you install and scope to all repositories or a selected set, or a saved personal access token. Source: https://docs.noriagentic.com/sessions/guides/integrations#github (checked 2026-11-06)
87. Amol Kapoor's Google-affiliated research includes MixHop (ICML 2019), N-GCN and a 2020 spatio-temporal graph neural network for COVID-19 forecasting. Source: https://scholar.google.com/citations?user=SC_-qB8AAAAJ (checked 2026-11-06)
88. Clifford Ressel is Nori's CTO and co-founder. Source: https://ai.engineer/orgs/nori (checked 2026-11-06)
89. Corvus's CTO saw a production bug at a client dinner, had Nori look into it and deploy a fix. Source: https://noriagentic.com/newsletter/2026-05-19-corvus-case-study.html (checked 2026-11-06)
90. Nori handles more than 25% of Miravoice's merged pull requests, and Miravoice's CTO says it easily saves him five hours a week. Source: https://noriagentic.com/newsletter/2026-09-18-miravoice-case-study.html (checked 2026-11-06)

---

Jiro is a background coding agent by Nori. It is not Amazon's Kiro or Atlassian's Jira.
