Gemini 3.5 Flash

Google

1049K context$1.50/M input$9.00/M output

1664peak 1665

Avg Score

79.8

Avg Cost

$0.27

Score/$

292.9

Runs

Win/Loss/Draw

Scoring Dimensions

Score Distribution

Category ELOs

from-scratchmedium

2251

backendeasy

2173

from-scratcheasy

2172

frontendexpert

2172

multi-languageexpert

2062

code-reviewhard

2018

from-scratchhard

1991

from-scratchexpert

1963

frontendhard

1879

from-scratch

1854

refactoringexpert

1806

refactoringmedium

1797

frontendeasy

1796

backendhard

1784

backendexpert

1777

debuggingmedium

1767

refactoring

1742

full-stackhard

1741

backend

1708

code-review

1690

code-reviewmedium

1686

full-stack

1667

backendmedium

1628

frontend

1587

full-stackmedium

1580

multi-language

1548

frontendmaster

1544

debugging

1535

multi-languagehard

1531

debuggingexpert

1524

frontendmedium

1490

backendmaster

1485

debugginghard

1476

All Results

Task	Category	Score	Duration	Cost
Fix and extend Chrome browser extension	frontend	58.3	11m 25s	$0.46
Build interactive data visualization dashboard	frontend	76.4	5m 0s	$0.20
Write tests for untested legacy Flask service	code-review	84.7	2m 36s	$0.20
Fix auth bypass vulnerability	debugging	78.8	4m 35s	$0.04
Build multi-tool LLM agent runtime	backend	80.8	4m 25s	$0.09
Build REST API from scratch	from-scratch	88.2	1m 11s	$0.03
Add virtual scrolling to table rendering 5000 rows	frontend	72.5	2m 48s	$0.30
Add GraphQL layer over REST API	multi-language	73.2	2m 57s	$0.32
Debug and fix 6 broken database triggers and constraints	debugging	82.5	2m 36s	$0.02
Harden insecure Docker setup with 12 vulnerabilities	code-review	88.5	1m 37s	$0.02
Implement JWT auth middleware	backend	74.3	10m 6s	$0.05
Add WebSocket real-time updates	full-stack	85.0	2m 57s	$0.02
Build distributed node cluster with gossip protocol	from-scratch	82.5	3m 48s	$0.08
Fix React hydration mismatch	frontend	70.0	4m 57s	$0.25
Implement zero-trust API authentication layer	backend	80.5	4m 11s	$0.12
Build materialized view refresh pipeline for analytics	backend	79.6	5m 3s	$0.03
Fix memory leak in event handler	debugging	71.3	3m 2s	$0.26
Add rate limiting middleware	backend	83.9	1m 17s	$0.02
Optimize bloated React bundle under 500KB	frontend	83.2	2m 26s	$0.06
Add caching layer to eliminate slow SSR page loads	full-stack	82.0	2m 18s	$0.19
Write complex SQL report with window functions	backend	88.5	2m 11s	$0.06
Add i18n with locale routing to Next.js app	full-stack	81.8	2m 40s	$0.32
Optimize slow Postgres queries in Flask app	backend	87.5	2m 27s	$0.03
Find and patch all OWASP Top 10 vulnerabilities	debugging	63.5	10m 5s	$0.06
Fix 12 WCAG accessibility violations in checkout form	frontend	85.1	1m 44s	$0.28
Add slash commands and moderation to Discord bot	backend	81.7	1m 32s	$0.04
Build SaaS admin dashboard from scratch	from-scratch	77.2	3m 23s	$0.33
Zero-downtime schema migration	full-stack	81.0	1m 36s	$0.02
Add Redis caching layer to Express API	backend	85.2	2m 16s	$0.04
Fix broken GitHub Actions CI pipeline	debugging	90.9	53.1s	$0.01
Code review: identify security vulns	code-review	73.5	1m 9s	$0.03
Fix N+1 query in dashboard	backend	73.0	1m 46s	$0.04
Build LLM evaluation harness with structured grading	backend	77.3	2m 23s	$0.90
Fix data integrity bugs in denormalized e-commerce schema	debugging	85.6	1m 59s	$0.02
Debug race condition in worker pool	debugging	84.7	59.3s	<$0.01
Build MCP server for database management	backend	83.3	1m 38s	$0.04
Fix Node.js stream backpressure causing OOM on large files	backend	91.3	1m 3s	$0.03
Add streaming SSE endpoint for LLM chat	backend	82.4	1m 47s	$0.03
Fix deadlocking transaction patterns in Flask app	backend	83.0	2m 40s	$0.03
Write Kubernetes manifests for Node.js microservice	full-stack	86.0	1m 7s	$0.03
Replace console.log with structured logging	refactoring	82.4	1m 13s	$0.05
Write integration tests for payment flow	code-review	80.0	2m 7s	$0.07
Build 3D browser game with physics and multiplayer sync	frontend	76.0	2m 22s	$0.07
Migrate Express monolith to modular architecture	backend	67.9	5m 3s	$0.28
Implement multi-tenant row-level security in Postgres	backend	79.2	2m 5s	$0.08
Migrate callback-hell Express app to async/await	refactoring	84.8	2m 57s	$0.71
Build RAG pipeline with vector search	backend	79.7	2m 28s	$0.02
Find and fix 4 hidden backdoors in Flask app	debugging	90.2	1m 20s	$0.02
Convert React app to PWA with offline support	frontend	74.6	2m 22s	$0.16
Add Google OAuth2 login to Express app	full-stack	76.3	1m 25s	$0.04
Add cursor-based pagination to REST API	backend	73.5	40.2s	$0.02
Add retry logic and dead letter queue to Python task queue	backend	79.8	2m 10s	$0.03
Fix broken responsive layout	frontend	78.7	1m 23s	$0.03
Implement background job scheduler with persistence	backend	79.7	3m 29s	$0.08
Build real-time portfolio risk calculator	backend	72.7	2m 31s	$0.06
Add file upload with S3 presigned URLs	backend	86.5	2m 7s	$0.13
Fix flaky test suite	debugging	87.5	5m 3s	<$0.01
Refactor monolithic handler to CQRS	refactoring	70.0	7m 13s	$1.61
Port Python CLI to Rust	multi-language	65.3	9m 4s	$1.88
Fix race conditions in order matching engine	backend	87.3	3m 23s	$0.93
Build terminal UI dashboard	from-scratch	75.0	3m 20s	$0.52
Dockerize Node.js monorepo	full-stack	76.7	1m 47s	$0.32
Implement transformer inference engine with KV cache	from-scratch	83.2	3m 17s	$0.83
Remove AI slop and over-engineering from codebase	refactoring	83.4	1m 52s	$0.50
Implement Stripe webhook handler	backend	82.9	1m 31s	$0.23
Fix hallucination and context window bugs in RAG agent	backend	77.7	3m 6s	$0.63
Build codebase indexer for LLM context windows	from-scratch	74.9	7m 31s	$1.98
Build CLI tool with subcommands and config	from-scratch	82.0	4m 59s	$1.33
Build production website with auth and members area	frontend	76.3	4m 32s	$0.76
Split 1100-line god file into proper modules	refactoring	83.7	3m 26s	$0.59