Back to models
GPT OSS 120b
OpenRouter
131K context$0.04/M input$0.19/M output
1290peak 1378
Avg Score
61.9
Avg Cost
$0.02
Score/$
3269.9
Runs
65
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
code-reviewhard
1936backendmedium
1438debugginghard
1416debuggingmedium
1391backendhard
1390debugging
1380backend
1371multi-languageexpert
1352debuggingexpert
1328full-stackhard
1301multi-language
1239backendexpert
1239full-stack
1224frontendmedium
1218from-scratchhard
1194code-review
1180frontendhard
1165frontend
1158from-scratch
1154full-stackmedium
1023refactoringmedium
973refactoring
948from-scratchexpert
897code-reviewmedium
805backendeasy
764from-scratcheasy
568from-scratchmedium
400multi-languagehard
306frontendexpert
163frontendeasy
37refactoringexpert
0All Results
| Task | Category | Score |
|---|---|---|
| Implement JWT auth middleware | backend | 55.5 |
| Zero-downtime schema migration | full-stack | 63.8 |
| Optimize bloated React bundle under 500KB | frontend | 64.0 |
| Code review: identify security vulns | code-review | 30.3 |
| Write complex SQL report with window functions | backend | 65.5 |
| Build real-time portfolio risk calculator | backend | 53.4 |
| Find and patch all OWASP Top 10 vulnerabilities | debugging | 63.3 |
| Build SaaS admin dashboard from scratch | from-scratch | 44.8 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 82.4 |
| Port Python CLI to Rust | multi-language | 53.8 |
| Add file upload with S3 presigned URLs | backend | 73.8 |
| Fix hallucination and context window bugs in RAG agent | backend | 31.5 |
| Migrate callback-hell Express app to async/await | refactoring | 55.2 |
| Add Redis caching layer to Express API | backend | 78.0 |
| Debug race condition in worker pool | debugging | 78.3 |
| Implement transformer inference engine with KV cache | from-scratch | 70.7 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 78.8 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 61.2 |
| Add retry logic and dead letter queue to Python task queue | backend | 65.8 |
| Add slash commands and moderation to Discord bot | backend | 65.7 |
| Implement zero-trust API authentication layer | backend | 59.1 |
| Split 1100-line god file into proper modules | refactoring | 43.3 |
| Build distributed node cluster with gossip protocol | from-scratch | 47.8 |
| Add GraphQL layer over REST API | multi-language | 48.0 |
| Fix auth bypass vulnerability | debugging | 89.7 |
| Build CLI tool with subcommands and config | from-scratch | 63.6 |
| Build LLM evaluation harness with structured grading | backend | 38.5 |
| Fix broken GitHub Actions CI pipeline | debugging | 79.8 |
| Add i18n with locale routing to Next.js app | full-stack | 55.8 |
| Add cursor-based pagination to REST API | backend | 86.5 |
| Fix N+1 query in dashboard | backend | 64.0 |
| Build production website with auth and members area | frontend | 48.3 |
| Implement multi-tenant row-level security in Postgres | backend | 57.9 |
| Write integration tests for payment flow | code-review | 73.8 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 78.1 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 73.5 |
| Fix race conditions in order matching engine | backend | 63.0 |
| Build MCP server for database management | backend | 78.8 |
| Build materialized view refresh pipeline for analytics | backend | 49.9 |
| Build terminal UI dashboard | from-scratch | 33.0 |
| Add caching layer to eliminate slow SSR page loads | full-stack | 76.3 |
| Add streaming SSE endpoint for LLM chat | backend | 84.8 |
| Optimize slow Postgres queries in Flask app | backend | 70.0 |
| Implement background job scheduler with persistence | backend | 53.1 |
| Fix flaky test suite | debugging | 81.3 |
| Dockerize Node.js monorepo | full-stack | 58.8 |
| Convert React app to PWA with offline support | frontend | 58.6 |
| Fix memory leak in event handler | debugging | 84.1 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 41.5 |
| Add Google OAuth2 login to Express app | full-stack | 78.9 |
| Implement Stripe webhook handler | backend | 82.8 |
| Fix broken responsive layout | frontend | 50.0 |
| Build codebase indexer for LLM context windows | from-scratch | 40.9 |
| Remove AI slop and over-engineering from codebase | refactoring | 65.8 |
| Debug and fix 6 broken database triggers and constraints | debugging | 82.0 |
| Replace console.log with structured logging | refactoring | 50.0 |
| Add rate limiting middleware | backend | 63.8 |
| Write tests for untested legacy Flask service | code-review | 35.2 |
| Add WebSocket real-time updates | full-stack | 70.7 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 75.0 |
| Build RAG pipeline with vector search | backend | 45.4 |
| Fix React hydration mismatch | frontend | 77.2 |
| Refactor monolithic handler to CQRS | refactoring | 28.6 |
| Fix deadlocking transaction patterns in Flask app | backend | 33.3 |
| Build REST API from scratch | from-scratch | 68.0 |