Key takeaways
- Amp is Sourcegraph's coding agent — built on their code intelligence infrastructure
- Multi-model approach: uses Opus 4.6, GPT-5.2, and fast models for different tasks
- Opinionated design philosophy: features get killed if the team doesn't love them
- Free tier with $10/day grant; pay-as-you-go with no markup for individuals
FAQ
What is Amp?
Amp is Sourcegraph's frontier coding agent for terminal and editor, using multiple AI models optimized for different coding tasks.
How much does Amp cost?
Free with a $10/day grant (ad-supported). Then pay-as-you-go with no markup for individuals.
What models does Amp use?
Amp uses Claude Opus 4.6 for most tasks, GPT-5.2 Codex for deep reasoning, plus fast models for quick operations.
Who competes with Amp?
Claude Code, Cursor, Factory Droids, and Tembo for orchestration.
Executive Summary
Amp is Sourcegraph's entry into the AI coding agent market, built on their decade of code intelligence infrastructure. It positions itself as a "frontier coding agent" with a multi-model approach and opinionated design philosophy. The generous free tier ($10/day grant) and transparent pay-as-you-go pricing differentiate it from subscription-based competitors.
| Attribute | Value |
|---|---|
| Company | Sourcegraph |
| Founded | 2013 (Amp launched 2025) |
| Funding | $223M (Sourcegraph total) |
| Employees | ~300 (Sourcegraph) |
| Headquarters | San Francisco, CA |
Product Overview
Amp is a terminal-first coding agent with IDE extensions, built on Sourcegraph's code intelligence infrastructure. [1]
The core philosophy is opinionated design: "If we don't use and love a feature, we kill it." They recently removed Tab completion entirely, calling it "not part of the future we see." [2]
Key Capabilities
| Capability | Description |
|---|---|
| Multi-Model Routing | Automatically selects best model for each task |
| 200k Token Context | Full codebase awareness |
| Threads | Version-controlled conversation history (shareable links) |
| AGENTS.md Support | Project-specific guidance files |
| Three Modes | Smart, Rush, and Deep for different workloads |
Product Surfaces / Editions
| Surface | Description | Availability |
|---|---|---|
| Terminal CLI | Primary interface, installed via curl or npm | GA |
| JetBrains Extension | IntelliJ, WebStorm, GoLand support | GA |
| Neovim Plugin | For terminal purists | GA |
| Web Interface | Browser-based access | GA |
Technical Architecture
Amp operates in three modes: [2]
- Smart mode — Unconstrained state-of-the-art model use (Opus 4.6 primary)
- Rush mode — Faster, cheaper, for small well-defined tasks
- Deep mode — Extended thinking with GPT-5.2 Codex for complex problems
Key Technical Details
| Aspect | Detail |
|---|---|
| Deployment | Cloud-based with local CLI |
| Model(s) | Claude Opus 4.6, GPT-5.2 Codex, fast models [3] |
| Integrations | JetBrains, Neovim, Sourcegraph |
| Open Source | No (proprietary) |
Strengths
- Sourcegraph DNA — Built on proven code intelligence infrastructure
- Multi-model approach — Uses best model for each task type automatically
- Transparent pricing — No markup, pay actual model costs
- Opinionated quality — Willing to remove features that don't work well
- Strong user sentiment — Twitter users report preferring Amp over Claude Code and Cursor
- Generous free tier — $10/day grant enables real evaluation
Cautions
- Terminal-first — May alienate developers who prefer GUI-heavy workflows
- Breaking changes — "No backcompat, no legacy features" is a double-edged sword
- Sourcegraph dependency — Future tied to Sourcegraph's strategic direction
- Limited enterprise features — Less mature than Factory or Cursor for large teams
- Newer product — Less battle-tested than established alternatives
Pricing & Licensing
| Tier | Price | Includes |
|---|---|---|
| Free | $0 | $10/day grant, ad-supported |
| Pay-as-you-go | Variable | No markup on model costs |
| Enterprise | Custom | Via Sourcegraph sales |
Licensing model: Freemium with usage-based pricing
Hidden costs: None — transparent model cost pass-through for individuals
Competitive Positioning
Direct Competitors
| Competitor | Differentiation |
|---|---|
| Claude Code | Claude Code is Anthropic-only; Amp is multi-model |
| Cursor | Cursor is IDE-native; Amp is terminal-first with lighter editor support |
| Factory AI | Factory has enterprise Droids; Amp is developer-focused |
| Codex | Codex is OpenAI-only; Amp uses multiple providers |
When to Choose Amp Over Alternatives
- Choose Amp when: You want multi-model capability with transparent pricing and terminal-first workflow
- Choose Claude Code when: You're committed to Claude models and want official Anthropic support
- Choose Cursor when: You prefer a full AI IDE over terminal agent
- Choose Factory when: You need enterprise-grade Droids with workflow automation
Ideal Customer Profile
Best fit:
- Terminal-native developers who prefer CLI over GUI
- Engineers wanting latest models without subscription lock-in
- Developers who value opinionated, curated tool experiences
- Teams already using or trusting Sourcegraph
- Budget-conscious users who appreciate transparent pricing
Poor fit:
- Developers who prefer heavy IDE integration
- Teams needing enterprise features (SSO, compliance, audit logs)
- Users who want stable, backward-compatible tooling
- Organizations requiring single-vendor model relationships
Viability Assessment
| Factor | Assessment |
|---|---|
| Financial Health | Strong — Sourcegraph is well-funded ($223M) |
| Market Position | Challenger — newer entrant in coding agent space |
| Innovation Pace | Rapid — opinionated updates, quick iteration |
| Community/Ecosystem | Growing — Sourcegraph's developer community |
| Long-term Outlook | Positive — Sourcegraph has staying power |
Amp benefits from Sourcegraph's established code intelligence business and funding. The risk is whether the terminal-first, opinionated approach resonates broadly enough for market leadership.
Bottom Line
Amp brings Sourcegraph's code intelligence expertise to the AI agent race. The multi-model approach and transparent pricing are compelling, but the terminal-first philosophy and willingness to break features may limit adoption.
Recommended for: Terminal-native developers who want multi-model capability with transparent, usage-based pricing and appreciate opinionated tool design.
Not recommended for: Teams needing enterprise features, stable backward compatibility, or heavy IDE integration.
Outlook: Amp will continue evolving rapidly with Sourcegraph's backing.[4][5] Expect deeper code intelligence integration and potential enterprise tier expansion. The multi-model approach positions it well as the model landscape continues fragmenting.
Research by Ry Walker Research • methodology