Back to models
Gemini 3 Pro Preview
OpenRouter
1049K context$2.00/M input$12.00/M output
1549peak 1647
Avg Score
75.0
Avg Cost
$0.41
Score/$
184.9
Runs
61
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
frontendeasy
2378from-scratchexpert
2076backendeasy
2073from-scratchmedium
2044refactoringexpert
1935from-scratcheasy
1807frontendhard
1715debuggingmedium
1684full-stackhard
1681backendmedium
1629full-stack
1626from-scratch
1615backend
1595backendexpert
1583full-stackmedium
1576code-reviewmedium
1573backendhard
1568frontend
1567refactoring
1546frontendmedium
1538code-review
1535from-scratchhard
1511refactoringmedium
1492code-reviewhard
1481debugginghard
1469multi-languageexpert
1455debugging
1434debuggingexpert
1333multi-language
1271frontendexpert
1185multi-languagehard
396All Results
| Task | Category | Score |
|---|---|---|
| Find and fix 4 hidden backdoors in Flask app | debugging | 93.7 |
| Build materialized view refresh pipeline for analytics | backend | 72.7 |
| Build CLI tool with subcommands and config | from-scratch | 72.2 |
| Add Redis caching layer to Express API | backend | 82.8 |
| Add slash commands and moderation to Discord bot | backend | 82.5 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 69.9 |
| Migrate callback-hell Express app to async/await | refactoring | 58.7 |
| Fix race conditions in order matching engine | backend | 86.6 |
| Fix N+1 query in dashboard | backend | 68.7 |
| Implement Stripe webhook handler | backend | 84.6 |
| Add streaming SSE endpoint for LLM chat | backend | 82.5 |
| Fix React hydration mismatch | frontend | 74.6 |
| Build LLM evaluation harness with structured grading | backend | 82.7 |
| Fix broken responsive layout | frontend | 73.1 |
| Zero-downtime schema migration | full-stack | 84.6 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 82.3 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 79.2 |
| Build REST API from scratch | from-scratch | 72.3 |
| Dockerize Node.js monorepo | full-stack | 73.2 |
| Fix flaky test suite | debugging | 87.6 |
| Optimize slow Postgres queries in Flask app | backend | 78.0 |
| Build MCP server for database management | backend | 85.3 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 58.6 |
| Code review: identify security vulns | code-review | 89.3 |
| Build RAG pipeline with vector search | backend | 75.3 |
| Write tests for untested legacy Flask service | code-review | 63.0 |
| Build real-time portfolio risk calculator | backend | 74.5 |
| Replace console.log with structured logging | refactoring | 61.9 |
| Add rate limiting middleware | backend | 86.8 |
| Fix auth bypass vulnerability | debugging | 86.5 |
| Optimize bloated React bundle under 500KB | frontend | 75.6 |
| Add GraphQL layer over REST API | multi-language | 51.4 |
| Debug and fix 6 broken database triggers and constraints | debugging | 76.3 |
| Write integration tests for payment flow | code-review | 42.4 |
| Convert React app to PWA with offline support | frontend | 61.4 |
| Add WebSocket real-time updates | full-stack | 83.3 |
| Port Python CLI to Rust | multi-language | 57.6 |
| Build distributed node cluster with gossip protocol | from-scratch | 71.6 |
| Build codebase indexer for LLM context windows | from-scratch | 51.0 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 85.0 |
| Add i18n with locale routing to Next.js app | full-stack | 75.1 |
| Build terminal UI dashboard | from-scratch | 72.3 |
| Implement JWT auth middleware | backend | 75.5 |
| Split 1100-line god file into proper modules | refactoring | 88.0 |
| Implement multi-tenant row-level security in Postgres | backend | 76.2 |
| Remove AI slop and over-engineering from codebase | refactoring | 84.9 |
| Implement zero-trust API authentication layer | backend | 79.9 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 82.8 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 90.5 |
| Implement background job scheduler with persistence | backend | 73.6 |
| Implement transformer inference engine with KV cache | from-scratch | 87.7 |
| Fix hallucination and context window bugs in RAG agent | backend | 61.0 |
| Build production website with auth and members area | frontend | 63.5 |
| Build SaaS admin dashboard from scratch | from-scratch | 69.0 |
| Write complex SQL report with window functions | backend | 80.7 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 75.8 |
| Fix deadlocking transaction patterns in Flask app | backend | 60.5 |
| Add retry logic and dead letter queue to Python task queue | backend | 80.5 |
| Refactor monolithic handler to CQRS | refactoring | 78.4 |
| Fix memory leak in event handler | debugging | 55.2 |
| Debug race condition in worker pool | debugging | 86.7 |