Updated 2026-07-28

Claude vs GPT Pricing: Full Breakdown

Is Claude cheaper than GPT? It depends on the model, the tier, and whether your workload is input-heavy or output-heavy. This page compares every Claude and GPT model side by side, with real dollar examples so you can pick the right one.

Head-to-Head: All Claude vs GPT Models

Here is every current Claude and GPT model with its per-token pricing and context window. All prices are per 1 million tokens.

ModelProviderInput / 1MOutput / 1MContext
Anthropic (Claude)
Claude Fable 5Anthropic$10$501M
Claude Opus 4.8Anthropic$5$251M
Claude Sonnet 5Anthropic$2$101M
Claude Haiku 4.5Anthropic$1$5200K
OpenAI (GPT)
GPT-5.6 SolOpenAI$5$301.05M
GPT-5.5 ProOpenAI$30$180400K
GPT-5.6 TerraOpenAI$2.50$151.05M
GPT-5.5OpenAI$5$30400K
GPT-5.6 LunaOpenAI$1$61.05M

Price Per Word and Per Page

Token pricing can feel abstract. Here is what each model costs when you think in words and pages instead. One token is roughly 0.75 words, so 1 million tokens is about 750,000 words. A standard page is ~500 words (~667 tokens).

ModelInput / wordOutput / wordInput / pageOutput / page
Claude Fable 5$0.000013$0.000067$0.007$0.03
Claude Opus 4.8$0.000007$0.000033$0.003$0.02
Claude Sonnet 5$0.000003$0.000013$0.001$0.007
Claude Haiku 4.5$0.000001$0.000007$0.0007$0.003
GPT-5.6 Sol$0.000007$0.000040$0.003$0.02
GPT-5.5 Pro$0.000040$0.0002$0.02$0.12
GPT-5.6 Terra$0.000003$0.000020$0.002$0.01
GPT-5.5$0.000007$0.000040$0.003$0.02
GPT-5.6 Luna$0.000001$0.000008$0.0007$0.004

Tier-by-Tier Comparison

The fairest comparison matches models of the same tier. Here is how each Claude model stacks up against its GPT counterpart.

Frontier: Claude Fable 5 vs GPT-5.6 Sol

Input / 1M:$10 vs $5GPT is 2x cheaper
Output / 1M:$50 vs $30GPT is 1.7x cheaper
Context:1M vs 1.05M

Flagship: Claude Opus 4.8 vs GPT-5.6 Terra

Input / 1M:$5 vs $2.50GPT is 2x cheaper
Output / 1M:$25 vs $15GPT is 1.7x cheaper
Context:1M vs 1.05M

Input-Heavy vs Output-Heavy Workloads

Which provider wins depends on what you are doing. Input-heavy workloads (reading documents, analyzing code, processing data) are dominated by input token costs. Output-heavy workloads (writing articles, generating code, producing reports) are dominated by output costs.

Input-Heavy (Reading)

The cheapest Claude model for input is Claude Haiku 4.5 at $1 / 1M. The cheapest GPT model for input is GPT-5.6 Luna at $1 / 1M.

Same input price at the budget tier

Output-Heavy (Writing)

The cheapest Claude model for output is Claude Haiku 4.5 at $5 / 1M. The cheapest GPT model for output is GPT-5.6 Luna at $6 / 1M.

Claude wins on output: 1.2x cheaper

Real Dollar Examples

Abstract per-token prices are hard to reason about. Here are three concrete scenarios with real costs on every Claude and GPT model.

Summarizing 500 pages

333,500 input tokens (500 pages at ~667 tokens/page)

  • Claude Fable 5: $3.33
  • Claude Opus 4.8: $1.67
  • Claude Sonnet 5: $0.67
  • Claude Haiku 4.5: $0.33
  • GPT-5.6 Sol: $1.67
  • GPT-5.5 Pro: $10.01
  • GPT-5.6 Terra: $0.83
  • GPT-5.5: $1.67
  • GPT-5.6 Luna: $0.33

Writing 50,000 words

66,667 output tokens (50,000 words at ~1.33 tokens/word)

  • Claude Fable 5: $3.33
  • Claude Opus 4.8: $1.67
  • Claude Sonnet 5: $0.67
  • Claude Haiku 4.5: $0.33
  • GPT-5.6 Sol: $2.00
  • GPT-5.5 Pro: $12
  • GPT-5.6 Terra: $1.00
  • GPT-5.5: $2.00
  • GPT-5.6 Luna: $0.40

Code review: 10,000 lines

100,000 input tokens (10,000 lines at ~10 tokens/line)

  • Claude Fable 5: $1
  • Claude Opus 4.8: $0.50
  • Claude Sonnet 5: $0.20
  • Claude Haiku 4.5: $0.10
  • GPT-5.6 Sol: $0.50
  • GPT-5.5 Pro: $3
  • GPT-5.6 Terra: $0.25
  • GPT-5.5: $0.50
  • GPT-5.6 Luna: $0.10

The Bottom Line

At the frontier tier, Claude Fable 5 is significantly cheaper than GPT-5.5 Pro on both input and output, while also offering a larger context window (1M vs 400K tokens).

At the flagship tier, Claude Opus 4.8 and GPT-5.5 share the same input price ($5 / 1M), but Claude Opus 4.8 is cheaper on output ($25vs $15) and comes with a much larger context window.

For budget workloads, Claude Haiku 4.5 at $1 / $5 is the cheapest Claude option. OpenAI does not currently offer a budget model in the GPT family, so Claude has the advantage for cost-conscious projects.

Short version: if you want the lowest price and the biggest context window, Claude wins across the board. GPT-5.5 competes at the flagship tier on input price but costs more for output.

Calculate your exact cost

Plug in your token usage and see what it costs on every model, side by side.

Open the token calculator →