Table of Contents
Complete AI Services Comparison: ChatGPT, Gemini, Copilot, Claude, Cursor, and Devin
In recent years, AI models and AI agents, starting with ChatGPT, have exploded in popularity. This article focuses on six representative services, covering pricing structures, high billing incidents, features, and future trends. We've organized the key points to make it readable in about 10 minutes.
💡 What you'll learn from this article: We'll comprehensively explain pricing plans for each AI service, actual high billing incidents, features and strengths, and future pricing trends.
1. ChatGPT (OpenAI)
💰 Pricing Plans
- Free Plan: Basic model only
- Plus: $20/month (about ¥3,000), access to GPT-4/5
- Pro: $200/month (about ¥30,000), GPT-5 with fewer restrictions
- Enterprise: $30-60/user/month, unlimited usage and enhanced security
🔧 API Pricing
Pay-per-use (e.g., GPT-4 Turbo input $0.01/1K tokens)
⚠️ High Billing Incidents
- Accidental multiple requests from Google Apps Script → Over ¥900,000 in 10 days [August 2024]
- API key leakage resulting in unauthorized charges of about ¥160,000 [2024]
📈 Value & Impact
Business productivity increased by 31-123%, with ROI estimates of 56-166x.
2. Google Gemini
🎯 Features
- Multimodal: Integrated processing of text, images, and audio
- Up to 1 million token context: Excellent for long text processing
- Search integration: Strong integration with Google Search
💰 Pricing Plans
- Free Plan: With limitations
- Gemini Advanced: $20/month, access to Gemini 1.5 Pro
- For Workspace: Business ($20/user/month), Enterprise ($30/user/month)
🔧 API Pricing
Gemini 1.5 Pro: Input $0.007/1K, Output $0.021/1K tokens
✅ Benefits
Powerful business efficiency through Gmail and Spreadsheet integration.
3. GitHub Copilot
🎯 Features
- AI Pair Programmer: Real-time code completion within the editor
- Intelligent suggestions: Context-aware code generation
💰 Pricing Plans
- Individual: $10/month (annual payment $100)
- Business: $19/user/month
- Enterprise: $39/user/month
✅ Peace of Mind
Fixed pricing means no worries about high charges.
📈 Impact
Survey results show development time reduced by an average of 55%.
4. Claude (Anthropic)
🎯 Features
- Up to 100k token input: Excellent for long text processing
- Polite writing style: Natural and readable responses
- Safety-focused: Promotes ethical AI development
💰 Pricing Plans
- Claude Pro: $20/month (5x usage limit)
- Claude Team: $30/user/month (200k token support)
🔧 API Pricing
Claude 2 (100k): Input $0.011/1K, Output $0.032/1K tokens
⚠️ Caution
Large inputs can consume several dollars.
✅ Strengths
Overwhelming power in long text processing and summarization.
5. Cursor
🎯 Features
- AI-integrated development environment: Complete integration of editor and AI
- Agent functionality: Autonomous coding assistance
- Bug fixing support: Automatic error detection and fix suggestions
💰 Pricing Plans
- Hobby: Free
- Pro: $20/month (unlimited completions)
- Ultra: $200/month (20x Pro usage limit)
📊 Real Case Study
A user who previously spent ¥90,000/month was able to fix costs at ¥30,000/month with Ultra.
✅ Value
AI autonomously assists with coding, and multiple agents can be used in parallel.
6. Devin
🎯 Features
- Autonomous AI Engineer: Fully automated development process
- Slack-based operation: Automatic execution of design, development, and testing
- End-to-end development: Consistent handling from requirements to implementation
📊 Performance
Benchmark results show about 7x the performance of conventional AI, though failures are also common.
💰 Pricing Plans
- Devin 1.0: $500/month (expensive)
- Devin 2.0: Starting from $20/month (price reduction for general developers)
🚀 Potential
"AI development teams" become reality with multiple agents running simultaneously.
Summary: Future AI Utilization and Cost Awareness
📉 Price Trends
Rapid price reductions due to competition (GPT-4 series dropped over 80% in 16 months).
🎯 Selection Approach
- ChatGPT/Claude → Text and knowledge work
- Copilot/Cursor → Code work
- Devin → Autonomous development
💡 Cost as Investment
Even monthly costs of several thousand to tens of thousands of yen can yield returns of dozens of times through productivity and new business ventures.
Conclusion
🎯 There is no single correct answer. Combine multiple AIs and choose tools that match your purpose and budget.
It's important to flexibly select AI services according to your use case and budget. By understanding the characteristics of each service and combining multiple ones, you can achieve maximum effectiveness.

NEW NOVEL 2026/08/01
Clouded Glass
Polishing is not about force.
Volume two of The World Became Slightly Farther Away.Five stories that can also be read as a starting point.
View on Amazon
Jijoden.com
Your life is worth writing.
There is a truer self you can tell only to AI.Gather fragments of memory into a single story.
Take a LookRelated Articles
Can AI Become an Emotional Pause Button?
A practical look at how conversational AI can serve as a temporary pause before harmful emotional action, while clarifying its risks, safer prompts, and when human or medical support is necessary.
Ichiro Furutachi Could Conquer the AI Agent Era: Why Voice Prompting Can Become a Real Advantage
This article argues that uncompressed context can outperform tidy short prompts, explains how voice input can prevent premature compression of thought, and shows why companies should stop overlooking “talking engineers” in the AI era.
What Is GPT-5.3 Instant?
A structured explainer of GPT-5.3 Instant covering design goals, latency engineering, accuracy measurements, safety trade-offs, and practical positioning based on public disclosures.
GPT-5.2 Explained: New Features, Performance, and Pricing Guide
A comprehensive guide to GPT-5.2 released by OpenAI on December 11, 2025, covering Instant/Thinking/Pro modes, improvements from GPT-5.1, API pricing, and competitive analysis.
Is ChatGPT Still the Runaway Leader? Why Google Struggles to Catch Up in the AI Race
Analyzes the current positions of ChatGPT and Gemini through users, loyalty, business models, and organizational culture, and explains what Japan's market and Google would need for a comeback.