Back to models
Minimax M2.7 [NVFP4]
SGLang
197K context<$0.01/M input<$0.01/M output
1578peak 1733
Avg Score
79.4
Avg Cost
—
Score/$
—
Runs
46
Win/Loss/Draw
Scoring Dimensions
Score Distribution
Category ELOs
multi-languageexpert
2300refactoringexpert
2101from-scratcheasy
1851multi-language
1818from-scratchhard
1768backendeasy
1762full-stackhard
1690code-reviewmedium
1687refactoring
1671full-stack
1645debuggingexpert
1640refactoringmedium
1631backendhard
1609debugging
1600debugginghard
1593backendexpert
1590frontendmedium
1577code-review
1566backend
1560from-scratch
1546full-stackmedium
1529frontend
1513backendmedium
1505backendmaster
1394from-scratchmedium
1355frontendhard
1341code-reviewhard
1282frontendmaster
1109from-scratchexpert
897All Results
| Task | Category | Score |
|---|---|---|
| Migrate Express monolith to modular architecture | backend | 70.3 |
| Implement JWT auth middleware | backend | 86.7 |
| Build REST API from scratch | from-scratch | 88.7 |
| Write integration tests for payment flow | code-review | 70.3 |
| Build CLI tool with subcommands and config | from-scratch | 74.3 |
| Build codebase indexer for LLM context windows | from-scratch | 77.7 |
| Fix and extend Chrome browser extension | frontend | 52.0 |
| Add Redis caching layer to Express API | backend | 83.4 |
| Optimize slow Postgres queries in Flask app | backend | 84.2 |
| Implement multi-tenant row-level security in Postgres | backend | 78.8 |
| Build materialized view refresh pipeline for analytics | backend | 77.3 |
| Fix N+1 query in dashboard | backend | 74.7 |
| Fix React hydration mismatch | frontend | 85.8 |
| Split 1100-line god file into proper modules | refactoring | 84.0 |
| Remove AI slop and over-engineering from codebase | refactoring | 81.9 |
| Fix auth bypass vulnerability | debugging | 91.0 |
| Harden insecure Docker setup with 12 vulnerabilities | code-review | 86.5 |
| Write Kubernetes manifests for Node.js microservice | full-stack | 86.3 |
| Build multi-tool LLM agent runtime | backend | 79.9 |
| Fix hallucination and context window bugs in RAG agent | backend | 80.8 |
| Fix 12 WCAG accessibility violations in checkout form | frontend | 77.7 |
| Refactor monolithic handler to CQRS | refactoring | 82.1 |
| Write tests for untested legacy Flask service | code-review | 84.3 |
| Fix race conditions in order matching engine | backend | 85.8 |
| Build terminal UI dashboard | from-scratch | 64.9 |
| Build MCP server for database management | backend | 81.5 |
| Debug and fix 6 broken database triggers and constraints | debugging | 86.8 |
| Migrate callback-hell Express app to async/await | refactoring | 85.1 |
| Implement background job scheduler with persistence | backend | 68.3 |
| Convert React app to PWA with offline support | frontend | 74.3 |
| Implement zero-trust API authentication layer | backend | 73.5 |
| Implement transformer inference engine with KV cache | from-scratch | 70.9 |
| Build RAG pipeline with vector search | backend | 76.5 |
| Write complex SQL report with window functions | backend | 82.4 |
| Fix Node.js stream backpressure causing OOM on large files | backend | 70.8 |
| Zero-downtime schema migration | full-stack | 81.5 |
| Find and fix 4 hidden backdoors in Flask app | debugging | 91.3 |
| Fix data integrity bugs in denormalized e-commerce schema | debugging | 85.2 |
| Add rate limiting middleware | backend | 80.8 |
| Replace console.log with structured logging | refactoring | 77.2 |
| Add Google OAuth2 login to Express app | full-stack | 77.7 |
| Optimize bloated React bundle under 500KB | frontend | 75.5 |
| Add virtual scrolling to table rendering 5000 rows | frontend | 81.8 |
| Port Python CLI to Rust | multi-language | 77.5 |
| Add cursor-based pagination to REST API | backend | 77.0 |
| Add WebSocket real-time updates | full-stack | 86.8 |