Back to models
GLM 4.7 Flash
OpenRouter
203K context$0.06/M input$0.40/M output
1277peak 1361
Avg Score
53.6
Avg Cost
$0.02
Score/$
2912.2
Runs
63
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
frontendeasy
1529debuggingmedium
1489frontendhard
1433debugginghard
1308debugging
1292backendmedium
1281full-stackmedium
1256code-reviewmedium
1247full-stack
1247frontend
1240code-review
1229backend
1212full-stackhard
1205refactoring
1184debuggingexpert
1172backendhard
1159from-scratcheasy
1150refactoringexpert
1147frontendexpert
1144backendexpert
1127refactoringmedium
1117from-scratch
1064frontendmedium
1051from-scratchmedium
992code-reviewhard
895from-scratchhard
881multi-language
835backendeasy
522from-scratchexpert
223multi-languagehard
121multi-languageexpert
32All Results
| Task | Category | Score |
|---|---|---|
| Fix hallucination and context window bugs in RAG agent | backend | 41.6 |
| Build distributed node cluster with gossip protocol | from-scratch | 29.4 |
| Port Python CLI to Rust | multi-language | 29.1 |
| Replace console.log with structured logging | refactoring | 56.1 |
| Debug and fix 6 broken database triggers and constraints | debugging | 74.5 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 54.0 |
| Fix flaky test suite | debugging | 37.5 |
| Add rate limiting middleware | backend | 43.6 |
| Build codebase indexer for LLM context windows | from-scratch | 14.8 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 37.6 |
| Optimize slow Postgres queries in Flask app | backend | 41.9 |
| Code review: identify security vulns | code-review | 80.2 |
| Write tests for untested legacy Flask service | code-review | 22.8 |
| Write complex SQL report with window functions | backend | 60.8 |
| Implement background job scheduler with persistence | backend | 40.7 |
| Zero-downtime schema migration | full-stack | 69.3 |
| Add GraphQL layer over REST API | multi-language | 44.3 |
| Add streaming SSE endpoint for LLM chat | backend | 81.8 |
| Add file upload with S3 presigned URLs | backend | 0.0 |
| Implement Stripe webhook handler | backend | 85.2 |
| Build CLI tool with subcommands and config | from-scratch | 38.0 |
| Add cursor-based pagination to REST API | backend | 77.0 |
| Add Redis caching layer to Express API | backend | 68.7 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 71.5 |
| Refactor monolithic handler to CQRS | refactoring | 40.7 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 57.6 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 52.5 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 80.5 |
| Migrate callback-hell Express app to async/await | refactoring | 30.7 |
| Remove AI slop and over-engineering from codebase | refactoring | 67.3 |
| Implement transformer inference engine with KV cache | from-scratch | 29.8 |
| Build production website with auth and members area | frontend | 61.9 |
| Build materialized view refresh pipeline for analytics | backend | 61.3 |
| Build MCP server for database management | backend | 57.4 |
| Optimize bloated React bundle under 500KB | frontend | 66.8 |
| Add i18n with locale routing to Next.js app | full-stack | 62.0 |
| Fix N+1 query in dashboard | backend | 26.9 |
| Add WebSocket real-time updates | full-stack | 63.0 |
| Implement JWT auth middleware | backend | 46.6 |
| Fix auth bypass vulnerability | debugging | 90.3 |
| Implement multi-tenant row-level security in Postgres | backend | 47.8 |
| Write integration tests for payment flow | code-review | 39.5 |
| Fix React hydration mismatch | frontend | 67.7 |
| Debug race condition in worker pool | debugging | 74.1 |
| Fix deadlocking transaction patterns in Flask app | backend | 68.3 |
| Build REST API from scratch | from-scratch | 61.1 |
| Build RAG pipeline with vector search | backend | 37.3 |
| Fix memory leak in event handler | debugging | 46.7 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 52.4 |
| Build LLM evaluation harness with structured grading | backend | 29.3 |
| Fix broken responsive layout | frontend | 77.5 |
| Build terminal UI dashboard | from-scratch | 58.5 |
| Split 1100-line god file into proper modules | refactoring | 57.8 |
| Implement zero-trust API authentication layer | backend | 31.0 |
| Convert React app to PWA with offline support | frontend | 35.8 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 78.1 |
| Dockerize Node.js monorepo | full-stack | 68.8 |
| Build real-time portfolio risk calculator | backend | 40.0 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 83.7 |
| Add slash commands and moderation to Discord bot | backend | 57.0 |
| Add retry logic and dead letter queue to Python task queue | backend | 40.0 |
| Build SaaS admin dashboard from scratch | from-scratch | 47.5 |
| Fix race conditions in order matching engine | backend | 82.2 |