Back to models
Qwen3.5 27b
OpenRouter
262K context$0.30/M input$2.40/M output
1421peak 1435
Avg Score
67.9
Avg Cost
$0.12
Score/$
563.4
Runs
70
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
refactoringexpert
2425multi-languagehard
2030code-reviewhard
1681frontendhard
1632debuggingmedium
1625full-stackhard
1620from-scratchexpert
1582backendmedium
1561debuggingexpert
1556debugging
1533debugginghard
1516multi-language
1513full-stack
1486backendexpert
1464backend
1439backendmaster
1433frontendexpert
1403frontendmaster
1393code-review
1388refactoring
1357code-reviewmedium
1323frontend
1309from-scratch
1302backendhard
1299full-stackmedium
1274from-scratcheasy
1249from-scratchhard
1198frontendmedium
1191refactoringmedium
1181from-scratchmedium
1039multi-languageexpert
934frontendeasy
761backendeasy
698All Results
| Task | Category | Score |
|---|---|---|
| Fix and extend Chrome browser extension | frontend | 32.0 |
| Build interactive data visualization dashboard | frontend | 73.0 |
| Build multi-tool LLM agent runtime | backend | 81.7 |
| Migrate Express monolith to modular architecture | backend | 65.7 |
| Build 3D browser game with physics and multiplayer sync | frontend | 77.8 |
| Build production website with auth and members area | frontend | 66.1 |
| Write integration tests for payment flow | code-review | 76.9 |
| Build SaaS admin dashboard from scratch | from-scratch | 73.0 |
| Implement background job scheduler with persistence | backend | 43.5 |
| Fix hallucination and context window bugs in RAG agent | backend | 35.2 |
| Implement multi-tenant row-level security in Postgres | backend | 37.5 |
| Write tests for untested legacy Flask service | code-review | 50.3 |
| Refactor monolithic handler to CQRS | refactoring | 84.3 |
| Optimize bloated React bundle under 500KB | frontend | 69.5 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 30.6 |
| Add WebSocket real-time updates | full-stack | 82.4 |
| Fix broken responsive layout | frontend | 62.3 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 80.5 |
| Build distributed node cluster with gossip protocol | from-scratch | 40.5 |
| Add cursor-based pagination to REST API | backend | 82.0 |
| Build materialized view refresh pipeline for analytics | backend | 72.7 |
| Build CLI tool with subcommands and config | from-scratch | 38.8 |
| Fix flaky test suite | debugging | 87.5 |
| Fix broken GitHub Actions CI pipeline | debugging | 85.5 |
| Implement zero-trust API authentication layer | backend | 71.8 |
| Remove AI slop and over-engineering from codebase | refactoring | 76.1 |
| Add i18n with locale routing to Next.js app | full-stack | 69.3 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 81.1 |
| Build real-time portfolio risk calculator | backend | 50.0 |
| Add Redis caching layer to Express API | backend | 84.9 |
| Fix race conditions in order matching engine | backend | 87.6 |
| Debug and fix 6 broken database triggers and constraints | debugging | 88.8 |
| Zero-downtime schema migration | full-stack | 82.2 |
| Add retry logic and dead letter queue to Python task queue | backend | 76.6 |
| Dockerize Node.js monorepo | full-stack | 67.2 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 67.8 |
| Fix memory leak in event handler | debugging | 80.1 |
| Add file upload with S3 presigned URLs | backend | 80.8 |
| Write complex SQL report with window functions | backend | 60.9 |
| Optimize slow Postgres queries in Flask app | backend | 81.0 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 75.0 |
| Build terminal UI dashboard | from-scratch | 56.0 |
| Replace console.log with structured logging | refactoring | 53.8 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 87.0 |
| Build LLM evaluation harness with structured grading | backend | 78.8 |
| Add Google OAuth2 login to Express app | full-stack | 80.7 |
| Fix auth bypass vulnerability | debugging | 90.8 |
| Build codebase indexer for LLM context windows | from-scratch | 28.8 |
| Fix N+1 query in dashboard | backend | 61.2 |
| Add streaming SSE endpoint for LLM chat | backend | 82.7 |
| Migrate callback-hell Express app to async/await | refactoring | 41.0 |
| Build MCP server for database management | backend | 51.4 |
| Build RAG pipeline with vector search | backend | 47.9 |
| Add GraphQL layer over REST API | multi-language | 83.8 |
| Add rate limiting middleware | backend | 46.3 |
| Implement transformer inference engine with KV cache | from-scratch | 77.5 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 70.0 |
| Build REST API from scratch | from-scratch | 74.6 |
| Code review: identify security vulns | code-review | 77.0 |
| Fix deadlocking transaction patterns in Flask app | backend | 72.3 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 79.3 |
| Implement JWT auth middleware | backend | 53.1 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 81.5 |
| Fix React hydration mismatch | frontend | 58.0 |
| Debug race condition in worker pool | debugging | 86.6 |
| Implement Stripe webhook handler | backend | 79.5 |
| Split 1100-line god file into proper modules | refactoring | 60.8 |
| Add slash commands and moderation to Discord bot | backend | 65.0 |
| Convert React app to PWA with offline support | frontend | 71.2 |
| Port Python CLI to Rust | multi-language | 43.8 |