Back to models
Qwen3.5 35b A3b
OpenRouter
262K context$0.25/M input$2.00/M output
1445peak 1456
Avg Score
63.4
Avg Cost
$0.08
Score/$
831.6
Runs
121
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
multi-languagehard
2051debuggingmedium
1792code-reviewmedium
1598backendexpert
1587code-review
1571frontendhard
1562full-stackmedium
1561code-reviewhard
1533full-stack
1529debugging
1523debuggingexpert
1523full-stackhard
1516backendeasy
1509frontendexpert
1471refactoring
1460backendmedium
1458refactoringmedium
1450frontendmedium
1440frontend
1434multi-language
1428from-scratchexpert
1414backend
1404debugginghard
1401frontendeasy
1315from-scratcheasy
1290from-scratchhard
1266from-scratch
1260backendhard
1247refactoringexpert
1156from-scratchmedium
1093multi-languageexpert
1020All Results
| Task | Category | Score |
|---|---|---|
| Build MCP server for database management | backend | 70.2 |
| Convert React app to PWA with offline support | frontend | 62.9 |
| Build real-time portfolio risk calculator | backend | 46.5 |
| Build CLI tool with subcommands and config | from-scratch | 30.0 |
| Add i18n with locale routing to Next.js app | full-stack | 46.1 |
| Split 1100-line god file into proper modules | refactoring | 34.5 |
| Write complex SQL report with window functions | backend | 59.7 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 77.8 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 61.1 |
| Add retry logic and dead letter queue to Python task queue | backend | 76.5 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 85.7 |
| Fix N+1 query in dashboard | backend | 56.5 |
| Fix race conditions in order matching engine | backend | 84.9 |
| Fix hallucination and context window bugs in RAG agent | backend | 30.3 |
| Fix memory leak in event handler | debugging | 81.8 |
| Debug race condition in worker pool | debugging | 81.1 |
| Fix broken GitHub Actions CI pipeline | debugging | 81.3 |
| Add streaming SSE endpoint for LLM chat | backend | 82.5 |
| Fix React hydration mismatch | frontend | 78.2 |
| Build LLM evaluation harness with structured grading | backend | 63.5 |
| Implement Stripe webhook handler | backend | 68.3 |
| Build distributed node cluster with gossip protocol | from-scratch | 26.2 |
| Migrate callback-hell Express app to async/await | refactoring | 62.4 |
| Build SaaS admin dashboard from scratch | from-scratch | 58.3 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 60.4 |
| Debug and fix 6 broken database triggers and constraints | debugging | 83.3 |
| Add Google OAuth2 login to Express app | full-stack | 75.4 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 86.7 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 77.5 |
| Implement background job scheduler with persistence | backend | 26.1 |
| Fix auth bypass vulnerability | debugging | 87.0 |
| Optimize slow Postgres queries in Flask app | backend | 78.0 |
| Build REST API from scratch | from-scratch | 60.9 |
| Add Redis caching layer to Express API | backend | 76.6 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 80.5 |
| Build materialized view refresh pipeline for analytics | backend | 71.0 |
| Optimize bloated React bundle under 500KB | frontend | 44.4 |
| Add file upload with S3 presigned URLs | backend | 71.9 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 77.5 |
| Fix flaky test suite | debugging | 86.6 |
| Build RAG pipeline with vector search | backend | 39.0 |
| Build codebase indexer for LLM context windows | from-scratch | 35.5 |
| Implement multi-tenant row-level security in Postgres | backend | 53.0 |
| Build terminal UI dashboard | from-scratch | 46.5 |
| Add cursor-based pagination to REST API | backend | 80.5 |
| Write integration tests for payment flow | code-review | 64.1 |
| Implement zero-trust API authentication layer | backend | 69.3 |
| Dockerize Node.js monorepo | full-stack | 59.4 |
| Add WebSocket real-time updates | full-stack | 65.3 |
| Add rate limiting middleware | backend | 70.5 |
| Implement transformer inference engine with KV cache | from-scratch | 73.8 |
| Implement JWT auth middleware | backend | 50.0 |
| Add GraphQL layer over REST API | multi-language | 82.5 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 76.5 |
| Remove AI slop and over-engineering from codebase | refactoring | 77.8 |
| Write tests for untested legacy Flask service | code-review | 42.6 |
| Build production website with auth and members area | frontend | 57.1 |
| Zero-downtime schema migration | full-stack | 85.8 |
| Replace console.log with structured logging | refactoring | 78.1 |
| Add slash commands and moderation to Discord bot | backend | 63.3 |
| Fix deadlocking transaction patterns in Flask app | backend | 76.7 |
| Fix broken responsive layout | frontend | 64.2 |
| Code review: identify security vulns | code-review | 34.1 |
| Port Python CLI to Rust | multi-language | 29.3 |
| Refactor monolithic handler to CQRS | refactoring | 37.6 |
| Port Python CLI to Rust | multi-language | 29.3 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 68.9 |
| Code review: identify security vulns | code-review | 79.0 |
| Add slash commands and moderation to Discord bot | backend | 61.8 |
| Add WebSocket real-time updates | full-stack | 71.8 |
| Implement multi-tenant row-level security in Postgres | backend | 69.5 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 62.5 |
| Optimize bloated React bundle under 500KB | frontend | 67.1 |
| Remove AI slop and over-engineering from codebase | refactoring | 73.6 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 86.7 |
| Add streaming SSE endpoint for LLM chat | backend | 74.7 |
| Build codebase indexer for LLM context windows | from-scratch | 31.0 |
| Implement zero-trust API authentication layer | backend | 66.5 |
| Convert React app to PWA with offline support | frontend | 67.8 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 68.8 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 89.2 |
| Split 1100-line god file into proper modules | refactoring | 67.0 |
| Fix broken responsive layout | frontend | 60.5 |
| Add i18n with locale routing to Next.js app | full-stack | 61.5 |
| Implement JWT auth middleware | backend | 76.5 |
| Replace console.log with structured logging | refactoring | 62.3 |
| Dockerize Node.js monorepo | full-stack | 74.8 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 77.1 |
| Write complex SQL report with window functions | backend | 75.6 |
| Build real-time portfolio risk calculator | backend | 28.2 |
| Build LLM evaluation harness with structured grading | backend | 61.0 |
| Build SaaS admin dashboard from scratch | from-scratch | 35.0 |
| Optimize slow Postgres queries in Flask app | backend | 74.5 |
| Debug and fix 6 broken database triggers and constraints | debugging | 87.7 |
| Fix hallucination and context window bugs in RAG agent | backend | 34.0 |
| Fix N+1 query in dashboard | backend | 50.2 |
| Add rate limiting middleware | backend | 31.4 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 73.2 |
| Write tests for untested legacy Flask service | code-review | 55.6 |
| Fix memory leak in event handler | debugging | 70.7 |
| Build CLI tool with subcommands and config | from-scratch | 47.9 |
| Build MCP server for database management | backend | 57.1 |
| Build RAG pipeline with vector search | backend | 29.4 |
| Refactor monolithic handler to CQRS | refactoring | 69.2 |
| Zero-downtime schema migration | full-stack | 61.9 |
| Build terminal UI dashboard | from-scratch | 32.0 |
| Fix auth bypass vulnerability | debugging | 77.6 |
| Implement background job scheduler with persistence | backend | 27.6 |
| Implement transformer inference engine with KV cache | from-scratch | 55.6 |
| Fix React hydration mismatch | frontend | 73.3 |
| Write integration tests for payment flow | code-review | 77.0 |
| Build distributed node cluster with gossip protocol | from-scratch | 21.1 |
| Fix race conditions in order matching engine | backend | 79.2 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 51.8 |
| Build production website with auth and members area | frontend | 62.4 |
| Build materialized view refresh pipeline for analytics | backend | 66.2 |
| Add retry logic and dead letter queue to Python task queue | backend | 69.1 |
| Fix deadlocking transaction patterns in Flask app | backend | 88.2 |
| Fix flaky test suite | debugging | 88.9 |
| Build REST API from scratch | from-scratch | 76.8 |
| Debug race condition in worker pool | debugging | 66.9 |