← Back to blog

How Much Does ChatGPT API Cost? OpenAI Pricing Guide 2026

OpenAI's API lineup now spans 9 models — from GPT-5.5 Pro at a staggering $180 per 1M output tokens to GPT-oss 20B at just $0.08 per 1M input tokens. Choosing the wrong model can mean a 2,000x difference in your bill.

This guide breaks down exactly what the ChatGPT API costs across every OpenAI model, with real-world scenarios and a free calculator to estimate your monthly spend.

OpenAI API Pricing at a Glance

Model Input (per 1M tokens) Output (per 1M tokens) Context Window Tier
GPT-5.5 Pro $30.00 $180.00 1M Ultra-Premium
GPT-5.5 $5.00 $30.00 1M Premium
GPT-5.3 Codex $1.75 $14.00 400K Mid
GPT-5 $1.25 $10.00 272K Mid
GPT-4o $2.50 $10.00 128K Mid
GPT-5 mini $0.25 $2.00 272K Budget
GPT-4o mini $0.15 $0.60 128K Budget
GPT-oss 120B $0.15 $0.60 128K Budget (Open Source)
GPT-oss 20B $0.08 $0.35 128K Budget (Open Source)

Key insight: GPT-4o is actually more expensive than GPT-5 on input tokens ($2.50 vs $1.25). If you're still using GPT-4o, switching to GPT-5 saves 50% on input costs with better performance.

Real-World Cost Examples

Scenario 1: AI Chatbot (1,000 messages/day)

Average: 1,500 input tokens, 500 output tokens per message. 30 days/month.

Monthly Chatbot Cost

GPT-5.5 $675.00/mo
GPT-5 $112.50/mo
GPT-4o $165.00/mo
GPT-5 mini $41.25/mo
GPT-4o mini $15.75/mo
GPT-oss 20B $8.85/mo
Claude Haiku 4.5 (comparison) $45.00/mo
Gemini 2.0 Flash (comparison) $6.00/mo

Key insight: GPT-4o mini at $15.75/mo is 43x cheaper than GPT-5.5 for chatbot workloads. For most chatbots, the budget tier is all you need.

Scenario 2: Code Generation (200 requests/day)

Average: 3,000 input tokens, 1,200 output tokens per request. 30 days/month.

Monthly Code Generation Cost

GPT-5.5 $2,250.00/mo
GPT-5.3 Codex $648.00/mo
GPT-5 $414.00/mo
GPT-4o $498.00/mo
GPT-5 mini $118.80/mo
GPT-4o mini $52.20/mo

Key insight: For code generation, GPT-5 ($414/mo) is cheaper than GPT-4o ($498/mo) while offering better code quality. The Codex variant costs more but is optimized for code tasks.

Scenario 3: RAG Pipeline (500 queries/day)

Average: 5,000 input tokens (context + query), 800 output tokens per query. 30 days/month.

Monthly RAG Cost

GPT-5.5 $810.00/mo
GPT-5 $285.00/mo
GPT-4o $345.00/mo
GPT-5 mini $61.50/mo
GPT-4o mini $28.50/mo
GPT-oss 120B $26.10/mo

Scenario 4: Document Summarization (100 documents/day)

Average: 10,000 input tokens, 500 output tokens per document. 30 days/month.

Monthly Summarization Cost

GPT-5.5 $600.00/mo
GPT-5 $187.50/mo
GPT-4o $225.00/mo
GPT-5 mini $45.00/mo
GPT-4o mini $21.00/mo

The Hidden Cost: Output Token Multipliers

Output tokens are 2x to 6x more expensive than input tokens across all OpenAI models. This is where costs really add up:

Model Input Price Output Price Output Multiplier
GPT-5.5 Pro $30.00 $180.00 6x
GPT-5.5 $5.00 $30.00 6x
GPT-5.3 Codex $1.75 $14.00 8x
GPT-5 $1.25 $10.00 8x
GPT-5 mini $0.25 $2.00 8x
GPT-4o $2.50 $10.00 4x
GPT-4o mini $0.15 $0.60 4x

What this means: Setting max_tokens is critical. An unbounded GPT-5.5 request generating 4,000 output tokens costs $0.12 in output alone — more than 100 GPT-4o mini requests.

OpenAI vs Competitors: Which is Cheapest?

Use Case Best OpenAI Model OpenAI Cost Cheapest Alternative Alternative Cost
Budget Chatbot GPT-4o mini $15.75/mo Gemini 2.0 Flash $6.00/mo
Code Generation GPT-5 $414.00/mo DeepSeek V4 Pro $96.60/mo
RAG Pipeline GPT-5 mini $61.50/mo Gemini 2.0 Flash $12.00/mo
Premium Reasoning GPT-5.5 $675.00/mo Claude Opus 4.8 $487.50/mo
Massive Context GPT-5.5 (1M) $675.00/mo Gemini 2.5 Pro (1M) $487.50/mo

OpenAI is rarely the cheapest option for any single use case. Its advantage is the broadest model ecosystem — 9 models from $0.08 to $180/1M tokens — so you can mix and match by workload.

How to Calculate Your ChatGPT API Costs

Cost Formula

Monthly Cost = (Input Tokens × Input Price + Output Tokens × Output Price) × Requests per Month ÷ 1,000,000

Example: 500 requests/day × 2,000 input tokens × $1.25/1M + 500 × 500 output × $10.00/1M = $37.50 input + $75 output = $112.50/month (GPT-5)

Or skip the math — use the APIpulse GPT-5 Cost Calculator to compare all OpenAI models side by side with Claude, Gemini, and DeepSeek.

5 Ways to Reduce Your OpenAI API Bill

  1. Ditch GPT-4o for GPT-5. GPT-5 costs 50% less on input ($1.25 vs $2.50) with better performance. If you're still on GPT-4o, switching is a free upgrade.
  2. Use GPT-4o mini for simple tasks. Classification, extraction, formatting — GPT-4o mini at $0.15/$0.60 handles these at 97% less cost than GPT-5.5.
  3. Set max_tokens religiously. Output tokens cost 4-8x more than input. Setting max_tokens to 500 instead of leaving it unbounded can cut costs 60%.
  4. Implement model routing. Route simple queries to GPT-4o mini, moderate to GPT-5, and only complex reasoning to GPT-5.5. This can reduce costs 40-70%.
  5. Consider GPT-oss models. GPT-oss 120B ($0.15/$0.60) matches GPT-4o mini pricing but with a 120B parameter open-source model. Good for self-hosted or batch workloads.

The Bottom Line

Most developers overpay for OpenAI API. If you're using GPT-5.5 for everything, you're spending 43x more than necessary for tasks that GPT-4o mini handles perfectly. The sweet spot for production workloads is GPT-5 ($1.25/$10) — it's cheaper than GPT-4o with better quality. Only use GPT-5.5 for tasks that genuinely need frontier reasoning.

Calculate your exact ChatGPT API costs. Enter your usage and compare with every alternative.

Try the Free GPT Calculator or Compare All Models

Want to optimize your AI API costs?

APIpulse Pro ($29 one-time) includes saved scenarios, cost report exports, and personalized recommendations that can save you up to 40%.

Get Pro — $29

Save money: APIpulse Cost Optimizer — find out how much you could save by switching models. Free tool.