Gemini 2.5 Pro

OpenRouter

1049K context$1.25/M input$10.00/M output

1484peak 1532

Avg Score

65.7

Avg Cost

$0.28

Score/$

237.9

Runs

Win/Loss/Draw

Scoring Dimensions

Score Distribution

Category ELOs

refactoringexpert

2017

from-scratchexpert

1909

from-scratchmedium

1888

frontendhard

1827

debuggingmedium

1785

from-scratcheasy

1736

backendeasy

1669

from-scratch

1611

from-scratchhard

1584

debugging

1562

debuggingexpert

1543

full-stackmedium

1543

debugginghard

1536

frontend

1519

frontendmedium

1515

backendhard

1499

refactoring

1494

full-stack

1473

backendmedium

1471

backend

1447

refactoringmedium

1414

frontendeasy

1414

full-stackhard

1393

multi-language

1354

code-reviewmedium

1344

frontendexpert

1337

code-review

1319

multi-languagehard

1305

backendexpert

1298

multi-languageexpert

1248

code-reviewhard

1074

All Results

Task	Category	Score	Duration	Cost
Build SaaS admin dashboard from scratch	from-scratch	50.1	3m 40s	$0.32
Split 1100-line god file into proper modules	refactoring	62.7	1m 53s	$0.21
Implement JWT auth middleware	backend	52.0	2m 7s	$0.26
Convert React app to PWA with offline support	frontend	52.1	3m 30s	$0.39
Add file upload with S3 presigned URLs	backend	74.2	2m 50s	$0.33
Implement multi-tenant row-level security in Postgres	backend	66.7	2m 31s	$0.25
Code review: identify security vulns	code-review	73.0	1m 43s	$0.23
Build terminal UI dashboard	from-scratch	59.3	1m 43s	$0.12
Build production website with auth and members area	frontend	65.9	6m 50s	$0.57
Add retry logic and dead letter queue to Python task queue	backend	70.5	4m 33s	$0.32
Add slash commands and moderation to Discord bot	backend	78.9	1m 60s	$0.16
Dockerize Node.js monorepo	full-stack	75.5	1m 31s	$0.17
Fix 12 WCAG accessibility violations in checkout form	frontend	86.5	3m 38s	$0.35
Add virtual scrolling to table rendering 5000 rows	frontend	85.6	1m 33s	$0.22
Implement zero-trust API authentication layer	backend	68.8	3m 3s	$0.31
Write integration tests for payment flow	code-review	41.4	4m 8s	$0.43
Fix React hydration mismatch	frontend	38.5	3m 16s	$0.35
Add WebSocket real-time updates	full-stack	58.7	2m 4s	$0.27
Harden insecure Docker setup with 12 vulnerabilities	code-review	65.2	3m 10s	$0.34
Build distributed node cluster with gossip protocol	from-scratch	58.0	2m 41s	$0.21
Implement background job scheduler with persistence	backend	40.2	7m 9s	$0.55
Add caching layer to eliminate slow SSR page loads	full-stack	85.2	1m 45s	$0.22
Write complex SQL report with window functions	backend	40.0	6m 41s	$0.58
Fix hallucination and context window bugs in RAG agent	backend	63.8	6m 5s	$0.50
Fix N+1 query in dashboard	backend	69.7	2m 14s	$0.18
Zero-downtime schema migration	full-stack	48.1	3m 8s	$0.25
Build real-time portfolio risk calculator	backend	58.6	5m 33s	$0.52
Refactor monolithic handler to CQRS	refactoring	28.3	3m 32s	$0.36
Optimize bloated React bundle under 500KB	frontend	79.7	2m 27s	$0.25
Build CLI tool with subcommands and config	from-scratch	73.3	2m 18s	$0.20
Implement transformer inference engine with KV cache	from-scratch	82.6	2m 44s	$0.23
Fix broken GitHub Actions CI pipeline	debugging	88.8	1m 4s	$0.13
Add GraphQL layer over REST API	multi-language	34.6	3m 8s	$0.35
Implement Stripe webhook handler	backend	66.5	3m 6s	$0.34
Find and patch all OWASP Top 10 vulnerabilities	debugging	31.8	4m 45s	$0.49
Replace console.log with structured logging	refactoring	60.8	3m 18s	$0.37
Add streaming SSE endpoint for LLM chat	backend	82.4	1m 34s	$0.18
Fix race conditions in order matching engine	backend	79.1	3m 56s	$0.35
Build materialized view refresh pipeline for analytics	backend	72.8	2m 58s	$0.26
Fix Node.js stream backpressure causing OOM on large files	backend	43.4	2m 58s	$0.33
Build MCP server for database management	backend	82.4	1m 6s	$0.09
Build codebase indexer for LLM context windows	from-scratch	52.9	4m 33s	$0.36
Fix flaky test suite	debugging	93.0	1m 44s	$0.14
Find and fix 4 hidden backdoors in Flask app	debugging	90.9	41.6s	$0.06
Add i18n with locale routing to Next.js app	full-stack	75.7	4m 39s	$0.27
Add rate limiting middleware	backend	73.5	3m 12s	$0.24
Debug and fix 6 broken database triggers and constraints	debugging	88.8	1m 51s	$0.17
Write tests for untested legacy Flask service	code-review	33.4	5m 32s	$0.46
Optimize slow Postgres queries in Flask app	backend	86.3	1m 20s	$0.12
Fix auth bypass vulnerability	debugging	78.5	2m 13s	$0.28
Debug race condition in worker pool	debugging	88.0	49.1s	$0.06
Fix broken responsive layout	frontend	75.0	2m 18s	$0.24
Build RAG pipeline with vector search	backend	37.8	2m 58s	$0.25
Fix memory leak in event handler	debugging	34.3	3m 53s	$0.36
Migrate callback-hell Express app to async/await	refactoring	64.4	2m 60s	$0.28
Build LLM evaluation harness with structured grading	backend	68.3	3m 27s	$0.29
Add Redis caching layer to Express API	backend	50.5	4m 38s	$0.46
Remove AI slop and over-engineering from codebase	refactoring	75.3	1m 24s	$0.16
Port Python CLI to Rust	multi-language	52.2	4m 47s	$0.33
Build REST API from scratch	from-scratch	70.2	1m 30s	$0.11
Write Kubernetes manifests for Node.js microservice	full-stack	82.3	55.8s	$0.07
Fix deadlocking transaction patterns in Flask app	backend	65.5	2m 52s	$0.17
Fix data integrity bugs in denormalized e-commerce schema	debugging	82.2	1m 23s	$0.12
Add cursor-based pagination to REST API	backend	90.0	1m 26s	$0.12