Back to models
GLM 4.5 Air
Z.ai
131K context$0.20/M input$1.10/M output
1299peak 1317
Avg Score
62.1
Avg Cost
$0.03
Score/$
1946.4
Runs
65
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
refactoringexpert
2588from-scratchexpert
1793refactoring
1625refactoringmedium
1526code-reviewhard
1505backendeasy
1480debuggingmedium
1419backendmedium
1361debugging
1332debugginghard
1318backend
1302backendexpert
1280frontendmedium
1268debuggingexpert
1259full-stackmedium
1238full-stack
1229code-review
1227backendhard
1202from-scratch
1199full-stackhard
1190frontend
1187frontendeasy
1144code-reviewmedium
1116from-scratcheasy
1067multi-language
1044from-scratchhard
1012multi-languagehard
687from-scratchmedium
589multi-languageexpert
285frontendhard
213frontendexpert
0All Results
| Task | Category | Score |
|---|---|---|
| Write integration tests for payment flow | code-review | 72.9 |
| Build production website with auth and members area | frontend | 28.7 |
| Implement transformer inference engine with KV cache | from-scratch | 82.0 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 68.2 |
| Add retry logic and dead letter queue to Python task queue | backend | 71.4 |
| Add Redis caching layer to Express API | backend | 63.0 |
| Remove AI slop and over-engineering from codebase | refactoring | 88.7 |
| Implement Stripe webhook handler | backend | 72.2 |
| Optimize bloated React bundle under 500KB | frontend | 70.5 |
| Fix flaky test suite | debugging | 76.4 |
| Write tests for untested legacy Flask service | code-review | 35.0 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 59.0 |
| Build codebase indexer for LLM context windows | from-scratch | 39.5 |
| Add rate limiting middleware | backend | 73.0 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 79.4 |
| Build REST API from scratch | from-scratch | 71.2 |
| Build terminal UI dashboard | from-scratch | 48.8 |
| Add cursor-based pagination to REST API | backend | 71.5 |
| Optimize slow Postgres queries in Flask app | backend | 79.2 |
| Fix React hydration mismatch | frontend | 69.3 |
| Add WebSocket real-time updates | full-stack | 67.2 |
| Fix race conditions in order matching engine | backend | 71.5 |
| Add GraphQL layer over REST API | multi-language | 56.9 |
| Write complex SQL report with window functions | backend | 57.7 |
| Implement background job scheduler with persistence | backend | 27.8 |
| Add Google OAuth2 login to Express app | full-stack | 65.8 |
| Convert React app to PWA with offline support | frontend | 65.8 |
| Build CLI tool with subcommands and config | from-scratch | 41.5 |
| Add slash commands and moderation to Discord bot | backend | 69.4 |
| Fix N+1 query in dashboard | backend | 56.5 |
| Add i18n with locale routing to Next.js app | full-stack | 56.2 |
| Fix broken responsive layout | frontend | 67.4 |
| Debug and fix 6 broken database triggers and constraints | debugging | 50.4 |
| Fix auth bypass vulnerability | debugging | 78.0 |
| Zero-downtime schema migration | full-stack | 64.1 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 78.3 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 56.4 |
| Fix hallucination and context window bugs in RAG agent | backend | 25.1 |
| Refactor monolithic handler to CQRS | refactoring | 86.6 |
| Implement JWT auth middleware | backend | 50.9 |
| Build distributed node cluster with gossip protocol | from-scratch | 40.0 |
| Port Python CLI to Rust | multi-language | 31.3 |
| Build SaaS admin dashboard from scratch | from-scratch | 31.9 |
| Build real-time portfolio risk calculator | backend | 33.9 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 74.6 |
| Build LLM evaluation harness with structured grading | backend | 62.3 |
| Fix deadlocking transaction patterns in Flask app | backend | 65.8 |
| Add streaming SSE endpoint for LLM chat | backend | 79.2 |
| Dockerize Node.js monorepo | full-stack | 67.0 |
| Fix memory leak in event handler | debugging | 79.1 |
| Build MCP server for database management | backend | 20.1 |
| Build materialized view refresh pipeline for analytics | backend | 73.1 |
| Add file upload with S3 presigned URLs | backend | 79.8 |
| Build RAG pipeline with vector search | backend | 45.8 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 51.5 |
| Implement zero-trust API authentication layer | backend | 67.5 |
| Debug race condition in worker pool | debugging | 78.7 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 77.3 |
| Migrate callback-hell Express app to async/await | refactoring | 62.5 |
| Implement multi-tenant row-level security in Postgres | backend | 66.7 |
| Fix broken GitHub Actions CI pipeline | debugging | 84.3 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 70.0 |
| Replace console.log with structured logging | refactoring | 68.9 |
| Split 1100-line god file into proper modules | refactoring | 72.7 |
| Code review: identify security vulns | code-review | 37.8 |