Back to models
Step 3.5 Flash
OpenRouter
256K context$0.10/M input$0.30/M output
1386peak 1438
Avg Score
53.8
Avg Cost
$0.03
Score/$
1945.5
Runs
65
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
from-scratchmedium
1748backendeasy
1712backendexpert
1525debugginghard
1515full-stackmedium
1491full-stack
1452frontendmedium
1450backend
1424code-reviewmedium
1416full-stackhard
1413backendmedium
1413from-scratchhard
1391from-scratch
1383code-review
1371backendhard
1360frontend
1354debugging
1344refactoringmedium
1324refactoring
1312from-scratcheasy
1252frontendexpert
1211debuggingmedium
1153debuggingexpert
1142frontendeasy
1094code-reviewhard
1074from-scratchexpert
849multi-language
685refactoringexpert
672frontendhard
354multi-languagehard
74multi-languageexpert
0All Results
| Task | Category | Score |
|---|---|---|
| Fix race conditions in order matching engine | backend | 65.8 |
| Port Python CLI to Rust | multi-language | 4.0 |
| Build CLI tool with subcommands and config | from-scratch | 49.5 |
| Add cursor-based pagination to REST API | backend | 6.5 |
| Add Google OAuth2 login to Express app | full-stack | 5.8 |
| Build codebase indexer for LLM context windows | from-scratch | 46.3 |
| Add retry logic and dead letter queue to Python task queue | backend | 28.0 |
| Build real-time portfolio risk calculator | backend | 27.2 |
| Implement JWT auth middleware | backend | 28.0 |
| Fix N+1 query in dashboard | backend | 80.0 |
| Implement background job scheduler with persistence | backend | 34.5 |
| Fix auth bypass vulnerability | debugging | 85.0 |
| Add streaming SSE endpoint for LLM chat | backend | 63.6 |
| Write integration tests for payment flow | code-review | 61.0 |
| Implement Stripe webhook handler | backend | 78.5 |
| Fix broken GitHub Actions CI pipeline | debugging | 59.7 |
| Write tests for untested legacy Flask service | code-review | 63.5 |
| Add file upload with S3 presigned URLs | backend | 74.8 |
| Add rate limiting middleware | backend | 28.0 |
| Add WebSocket real-time updates | full-stack | 27.9 |
| Fix flaky test suite | debugging | 23.0 |
| Fix memory leak in event handler | debugging | 70.5 |
| Refactor monolithic handler to CQRS | refactoring | 52.0 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 28.0 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 28.0 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 28.0 |
| Code review: identify security vulns | code-review | 22.0 |
| Optimize bloated React bundle under 500KB | frontend | 51.8 |
| Debug and fix 6 broken database triggers and constraints | debugging | 41.8 |
| Migrate callback-hell Express app to async/await | refactoring | 22.0 |
| Add i18n with locale routing to Next.js app | full-stack | 28.0 |
| Optimize slow Postgres queries in Flask app | backend | 79.1 |
| Dockerize Node.js monorepo | full-stack | 28.0 |
| Fix broken responsive layout | frontend | 32.0 |
| Build LLM evaluation harness with structured grading | backend | 63.8 |
| Implement transformer inference engine with KV cache | from-scratch | 43.0 |
| Remove AI slop and over-engineering from codebase | refactoring | 22.0 |
| Fix React hydration mismatch | frontend | 62.9 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 90.9 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 59.1 |
| Build terminal UI dashboard | from-scratch | 69.7 |
| Add GraphQL layer over REST API | multi-language | 54.0 |
| Build MCP server for database management | backend | 76.9 |
| Build REST API from scratch | from-scratch | 72.7 |
| Add Redis caching layer to Express API | backend | 57.8 |
| Fix deadlocking transaction patterns in Flask app | backend | 79.2 |
| Convert React app to PWA with offline support | frontend | 84.8 |
| Build production website with auth and members area | frontend | 64.3 |
| Split 1100-line god file into proper modules | refactoring | 80.8 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 81.7 |
| Build RAG pipeline with vector search | backend | 43.0 |
| Implement multi-tenant row-level security in Postgres | backend | 74.8 |
| Build SaaS admin dashboard from scratch | from-scratch | 44.3 |
| Write complex SQL report with window functions | backend | 41.8 |
| Add slash commands and moderation to Discord bot | backend | 53.3 |
| Fix hallucination and context window bugs in RAG agent | backend | 50.5 |
| Implement zero-trust API authentication layer | backend | 77.3 |
| Debug race condition in worker pool | debugging | 87.9 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 81.7 |
| Replace console.log with structured logging | refactoring | 60.0 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 88.4 |
| Zero-downtime schema migration | full-stack | 82.5 |
| Build materialized view refresh pipeline for analytics | backend | 80.0 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 52.3 |
| Build distributed node cluster with gossip protocol | from-scratch | 66.8 |