Show HN: CostPerPrompt – Live AI API pricing and real-workload cost calculators

4 hours ago

AI pricing is based on tokens (roughly 3/4 of a word), with separate charges for input and output; output is typically 3-5x more expensive than input.
Prompt caching can reduce repeated input costs by up to 90%, and batch processing offers ~50% discounts, which many cost estimates ignore.
A live table shows pricing for 232+ flagship models, including GPT-5.x, Claude 5 variants, DeepSeek V4, Gemini 3.x, Grok 4.x, and GLM 5.2, with costs ranging from $0.09 to $10.00 per million input tokens.
Calculators are provided to convert token prices into real monthly costs for chatbots, agents, or API workloads.

AI cost in tokens and dollars is a major concern, with examples like Uber exceeding its budget quickly.
Coding agents involve different token types: input, cached input, output, and reasoning tokens, each with different prices and implications.
Cached input tokens are the cheapest but can dominate costs due to repeated use in agentic loops, while output tokens are the most expensive.

The author argues that AI inference is not as costly as commonly claimed, using H100 GPU pricing and DeepSeek R1 architecture as a baseline.
Input processing is extremely cheap (~$0.003 per million tokens) compared to output generation (~$3 per million tokens) due to sequential decode phase.
This cost asymmetry makes heavy-input, low-output use cases (e.g., coding assistants) highly profitable, with consumer plans showing 5-6x markup and developer plans up to 20x.

Comparing AI model costs by price per 1M tokens is meaningless due to differences in tokenizers and token efficiency.
Each frontier lab uses its own tokenizer, causing the same text to be split into varying token counts, making direct price comparisons unreliable.
Token efficiency—how much is achieved per token—varies widely, especially with 'thinking' tokens in chain-of-thought processes, significantly impacting overall costs.

Token counts for the same content vary significantly between AI models due to different tokenizers, which is not reflected in the advertised price per million tokens.
Anthropic's new tokenizer (used in Claude models like Opus 4.8 and Sonnet 5) increases token counts by about 30-32% compared to its old tokenizer, despite unchanged headline prices, leading to higher effective costs.
The tokenizer divergence is most pronounced for code, especially TypeScript, where Claude's new tokenizer uses 1.73x more tokens than OpenAI's GPT (with o200k tokenizer), compared to 1.4x for English prose.

Claude Code uses significantly more tokens than OpenCode for the same tasks, with baseline overhead around 33,000 tokens vs 7,000 for a simple reply.
Cache inefficiency is a major factor; Claude Code rewrites tens of thousands of prompt-cache tokens mid-session, leading to up to 54x more cache writes than OpenCode.
Configuration elements like instruction files and MCP servers add substantial token overhead, with a real setup reaching 75,000–85,000 tokens before user input.

AI cost in tokens and dollars is a major concern, with examples like Uber exceeding its budget quickly.
Coding agents involve different token types: input, cached input, output, and reasoning tokens, each with different prices and implications.
Cached input tokens are the cheapest but can dominate costs due to repeated use in agentic loops, while output tokens are the most expensive.

The author argues that AI inference is not as costly as commonly claimed, using H100 GPU pricing and DeepSeek R1 architecture as a baseline.
Input processing is extremely cheap (~$0.003 per million tokens) compared to output generation (~$3 per million tokens) due to sequential decode phase.
This cost asymmetry makes heavy-input, low-output use cases (e.g., coding assistants) highly profitable, with consumer plans showing 5-6x markup and developer plans up to 20x.

Comparing AI model costs by price per 1M tokens is meaningless due to differences in tokenizers and token efficiency.
Each frontier lab uses its own tokenizer, causing the same text to be split into varying token counts, making direct price comparisons unreliable.
Token efficiency—how much is achieved per token—varies widely, especially with 'thinking' tokens in chain-of-thought processes, significantly impacting overall costs.

Token counts for the same content vary significantly between AI models due to different tokenizers, which is not reflected in the advertised price per million tokens.
Anthropic's new tokenizer (used in Claude models like Opus 4.8 and Sonnet 5) increases token counts by about 30-32% compared to its old tokenizer, despite unchanged headline prices, leading to higher effective costs.
The tokenizer divergence is most pronounced for code, especially TypeScript, where Claude's new tokenizer uses 1.73x more tokens than OpenAI's GPT (with o200k tokenizer), compared to 1.4x for English prose.

Claude Code uses significantly more tokens than OpenCode for the same tasks, with baseline overhead around 33,000 tokens vs 7,000 for a simple reply.
Cache inefficiency is a major factor; Claude Code rewrites tens of thousands of prompt-cache tokens mid-session, leading to up to 54x more cache writes than OpenCode.
Configuration elements like instruction files and MCP servers add substantial token overhead, with a real setup reaching 75,000–85,000 tokens before user input.

Hasty Briefsbeta