Back to modelsGoogle
Gemini 3 Flash Preview
1049K context$0.50/M input$3.00/M output
1544peak 1559
Avg Score
72.0
Avg Cost
$0.02
Score/$
4126.4
Runs
46
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
refactoringexpert
2558from-scratchexpert
2072debuggingmedium
2010backendeasy
1964from-scratcheasy
1758frontendhard
1756from-scratchhard
1725full-stackhard
1699multi-languagehard
1695backendmedium
1684refactoring
1651from-scratch
1649full-stack
1629code-reviewmedium
1603backend
1541full-stackmedium
1526multi-language
1514code-review
1498frontend
1488frontendmedium
1488backendhard
1475debugging
1458backendexpert
1453refactoringmedium
1428debuggingexpert
1378debugginghard
1342frontendexpert
1136from-scratchmedium
1004multi-languageexpert
917code-reviewhard
188All Results
| Task | Category | Score |
|---|---|---|
| Build codebase indexer for LLM context windows | from-scratch | 60.7 |
| Optimize bloated React bundle under 500KB | frontend | 72.0 |
| Migrate callback-hell Express app to async/await | refactoring | 72.7 |
| Add WebSocket real-time updates | full-stack | 81.0 |
| Port Python CLI to Rust | multi-language | 51.3 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 88.8 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 84.5 |
| Split 1100-line god file into proper modules | refactoring | 68.5 |
| Implement transformer inference engine with KV cache | from-scratch | 87.7 |
| Build MCP server for database management | backend | 70.3 |
| Build SaaS admin dashboard from scratch | from-scratch | 75.0 |
| Implement background job scheduler with persistence | backend | 65.5 |
| Build production website with auth and members area | frontend | 55.6 |
| Build CLI tool with subcommands and config | from-scratch | 48.1 |
| Fix hallucination and context window bugs in RAG agent | backend | 55.5 |
| Write complex SQL report with window functions | backend | 71.9 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 71.5 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 84.8 |
| Fix race conditions in order matching engine | backend | 78.1 |
| Build real-time portfolio risk calculator | backend | 62.4 |
| Build LLM evaluation harness with structured grading | backend | 72.0 |
| Fix deadlocking transaction patterns in Flask app | backend | 44.3 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 78.4 |
| Debug and fix 6 broken database triggers and constraints | debugging | 66.6 |
| Add Redis caching layer to Express API | backend | 79.0 |
| Optimize slow Postgres queries in Flask app | backend | 86.3 |
| Add Google OAuth2 login to Express app | full-stack | 80.9 |
| Write tests for untested legacy Flask service | code-review | 47.3 |
| Add retry logic and dead letter queue to Python task queue | backend | 72.0 |
| Add GraphQL layer over REST API | multi-language | 73.8 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 90.4 |
| Build distributed node cluster with gossip protocol | from-scratch | 67.9 |
| Write integration tests for payment flow | code-review | 35.8 |
| Add rate limiting middleware | backend | 82.9 |
| Implement Stripe webhook handler | backend | 70.9 |
| Zero-downtime schema migration | full-stack | 73.0 |
| Fix flaky test suite | debugging | 91.2 |
| Add cursor-based pagination to REST API | backend | 69.0 |
| Fix N+1 query in dashboard | backend | 91.7 |
| Fix memory leak in event handler | debugging | 66.0 |
| Refactor monolithic handler to CQRS | refactoring | 80.3 |
| Code review: identify security vulns | code-review | 88.1 |
| Debug race condition in worker pool | debugging | 84.0 |
| Fix React hydration mismatch | frontend | 76.7 |
| Build terminal UI dashboard | from-scratch | 51.0 |
| Build REST API from scratch | from-scratch | 85.7 |