Compare/Agents
Compare AI agents
Pick up to 4 agents. Capability benchmarks, real-world OpenRouter usage, base models, license, and pricing render side-by-side.
Looking to compare base models instead? Switch to model compare →
| Comet Perplexity | Lovable Lovable | Claude Code Anthropic | Devin Cognition | |
|---|---|---|---|---|
| Overview | ||||
| Vendor | Perplexity | Lovable | Anthropic | Cognition |
| Category | browser | coding | coding | coding |
| Release date | — | — | — | — |
| Source | proprietary | proprietary | proprietary | proprietary |
| License | — | — | — | — |
| Pricing | Bundled with Perplexity Pro $20/mo | $20/mo Pro | Included with Claude Pro/Max/Team subscriptions | $500/mo Team |
| Base models | — | — | — | |
| Capabilities | browsersearchagentic | full_stackapp_generationauthdatabase | terminalfile_iomulti_file_editshell_exec+1 | autonomouscloud_runtimemulti_sessionbrowser_use |
| Capability | ||||
| gpt.buzz Index | — | — | 95.4 | 12.4 |
| SWE-bench Verified | — | — | 80.9% | 35.0% |
| SWE-bench Pro | — | — | 55.4% | — |
| Terminal-Bench 2.0 | — | — | 65.4% | — |
| GAIA | — | — | 74.6% | — |
| WebArena | — | — | 64.5% | — |
| Adoption | ||||
| OpenRouter · daily | — | — | 73.5B | — |
| OpenRouter · cumulative | — | — | 3.52T | — |
| GitHub stars | — | — | — | — |
| Links | ||||
| Links | ||||
Add an agent
Replit AgentReplit · codingmaxHermes AgentNous Research · generalmaxClaude CodeAnthropic · coding✓CodexOpenAI · codingmaxCursorAnysphere · codingmaxGitHub CopilotGitHub · codingmaxClineCline · codingmaxOpenCodeOpenCode · codingmaxDevinCognition · coding✓ManusMonica · generalmaxComputer UseAnthropic · browsermaxOperatorOpenAI · browsermaxCometPerplexity · browser✓v0Vercel · codingmaxBolt.newStackBlitz · codingmaxLovableLovable · coding✓OpenClawErik Steinberger · generalmax
Max 4 agents. Remove one to add another.