Amp
#16 of 21 · 8.0/10 editorialCoding AgentProprietary56% community win rate

Amp Review 2026: Ratings, Pros & Cons, Alternatives

Amp started inside Sourcegraph and became an independent company in December 2025. It stays deliberately opinionated: instead of a model picker it offers a four-step effort dial (low, medium, high, ultra), with Claude Opus 5.5 behind medium since September 28, 2026, and leans into autonomy, with subagents such as Oracle and Librarian for parallel work. Since September you can also pin the models behind each mode. Work happens in shareable threads across its CLI, web and desktop apps, plus cloud runners it calls orbs. Pricing is a free Hobby tier and a $20/month Individual plan that adds orb time; model usage bills from credits at providers' API prices with no markup. Every tier can also bring its own keys or endpoint through Model Routing, link a ChatGPT subscription, or, since October 10, 2026, run a Claude Code mode on a Claude Pro or Max plan. It is token-hungry by design, so heavy use still gets expensive.

Visit Amp →Pricing: Free Hobby tier / $20/mo IndividualUpdated 2026-10-16
What it costs to run Amp

4 options compared, with prices and limits.

Best API for Amp →See the Amp integration →

How Amp Scores

Output Quality
8.5/1059% votes
Autonomy
8.5/10
Reliability
8.0/10
Speed
8.0/10
Value
7.0/10
Ease of Use
8.0/10

Bars show editorial scores from hands-on testing. Percentages showAmp's live community win rate in head-to-head votes, shown once a category has 10 or more votes.

Amp Pros, Cons, and Known Issues

Pros

  • Opinionated defaults: an effort dial (low, medium, high, ultra) picks strong models for you, with Claude Opus 5.5 behind medium
  • Subagents such as Oracle and Librarian, and shareable threads across CLI, web and desktop
  • Model usage on Amp credits has no markup over providers' API prices
  • Model Routing accepts your own keys or any OpenAI- or Anthropic-compatible endpoint on every tier, free Hobby included
  • Can run on a ChatGPT subscription, or on a Claude Pro or Max plan in its Claude Code mode

Cons

  • Token-hungry by design, so heavy use on credits gets expensive
  • Closed source
  • Web search, image generation, voice and orb minutes stay on Amp credits even with your own endpoint

Known issues users report

  • A thread keeps its mode and models from the first message, so pin models before you start
  • Model mapping IDs must match what your provider serves; Amp writes Claude IDs with dashes
  • The Claude Code mode runs the Claude Agent SDK rather than Amp's own agent

Key Features

Effort dial instead of a model picker
Shareable threads & team visibility
Subagents for parallel work
CLI, web & desktop apps
Cloud runners (orbs)
Own keys or endpoint, ChatGPT or Claude plan

Runs on: Terminal (CLI) · Web · macOS app|Best for: Developers and teams who want an opinionated, high-autonomy agent and accept its token appetite

Amp vs the Alternatives: Vote in Any Matchup

AmpVSClaude CodeAmpVSHermesAmpVSCodexAmpVSGemini CLIAmpVSQwen CodeAmpVSOpenClawAmpVSKilo CodeAmpVSCursorAmpVSCopilotAmpVSClineAmpVSWindsurfAmpVSAiderAmpVSContinueAmpVSOpenCodeAmpVSPiAmpVSompAmpVSRoo CodeAmpVSDevinAmpVSJulesAmpVSAntigravity

More intelligence per dollar.

Amp runs on Standard Compute. Each step goes to the cheapest model that can do it, on providers we vet for price, uptime and speed, with automatic fallback when one fails. Current frontier models: Claude Fable 5.2 and GPT-6 Astra. Opt-in rules keep a company, team or key on European or US providers, and we don’t train on your prompts or store them.

Get your API keyBest API for Amp

Fixed monthly plans from $19 · cancel anytime · Amp setup

Amp: Frequently Asked Questions

Does Amp still let you bring your own API key?

Yes, as of October 2026. Amp's Model Routing docs list Custom URL connections (chat-completions, responses, anthropic-messages or google-genai formats) for all users on every tier, and the pricing page lists bring-your-own-key with no Amp token fees, including on the free Hobby plan. Only the models you map move to your provider.

Can I use my Claude Pro or Max plan in Amp?

Yes, since October 10, 2026. Pick Mode → Claude Code when you create a thread and link your Claude subscription. That mode runs Claude Opus 5.5 at high effort through the Claude Agent SDK instead of Amp's own agent, at no extra Amp cost.

How much does Amp cost?

Hobby is free with every product feature; Individual is $20 a month and adds 45,000 minutes of orb time. Model usage bills from credits at providers' API prices with no markup, and credits expire 12 months after purchase. Teams adds no extra charge; Enterprise is custom.