# Anthropic API

58 Hacker News commenters recommended Anthropic API and 126 criticized it, November 2025 to November 2026. It was recommended in 12 of 13 months.

**Verdict:** Recommended for straightforward pay-per-use access to Claude models with clear, stable pricing and no client lock-in. Common caveat: high token costs and uptime or account-suspension worries.

## Why people recommend it

- Straightforward and easy to use (2 comments): "$200 free credits from Anthropic are stupid simple to use" ([source](https://news.ycombinator.com/item?id=50079662)) "The API is there. It's straightforward and easy to use." ([source](https://news.ycombinator.com/item?id=47638268))
- Flexible access without client lock-in (3 comments): "The API gives you the same flexibility without the risk of getting your account suspended." ([source](https://news.ycombinator.com/item?id=46912922)) "They have a separate product intended for use with general tools: Their API." ([source](https://news.ycombinator.com/item?id=47447206)) "just use the API which doesn't lock you into any client" ([source](https://news.ycombinator.com/item?id=46550319))
- Stable costs and clear pricing (3 comments): "their API has a model selection anyone can understand and pricing that rarely changes" ([source](https://news.ycombinator.com/item?id=48336551)) "If you want stable costs, the API plan is right there." ([source](https://news.ycombinator.com/item?id=48805605)) "The API offers that. Pay X per month, get Y tokens." ([source](https://news.ycombinator.com/item?id=47638905))
- Direct developer API access (2 comments): "direct-to-dev type API access oAI and Anthropic have pioneered" ([source](https://news.ycombinator.com/item?id=46038003)) "If you don't want to use CC, just use the API" ([source](https://news.ycombinator.com/item?id=46554655))

## What people warn about

- Token costs high and unpredictable (4 comments): "It's pretty easy to hit a couple hundred dollars a day filling up Opus's context window with files." ([source](https://news.ycombinator.com/item?id=46296344)) "Anthropic is also so insanely expensive for everything" ([source](https://news.ycombinator.com/item?id=49548999)) "it will just start charging you API prices for the tokens you are generating without even a single warning" ([source](https://news.ycombinator.com/item?id=49997172))
- Data use and training concerns (3 comments): "Anthropic has made their arguments (that I disagree with) that model outputs, no matter how much user context was fed, do not fall under their definition of "user data"." ([source](https://news.ycombinator.com/item?id=50237519)) "when they say they store your data for 30 days and review it for "issues", it makes your "spider sense" tingle" ([source](https://news.ycombinator.com/item?id=48476151)) "The idea OpenAI or Anthropic won't train on your data—even with an enterprise contract—is a fantasy at best" ([source](https://news.ycombinator.com/item?id=49615154))
- Accounts and keys suspended without warning (2 comments): "Within seconds I got an email from Anthropic and they have already disabled my keys." ([source](https://news.ycombinator.com/item?id=46251507)) "Anthropic is unsurprisingly leading the race to the bottom of the barrel by requiring moderator reviews for API customers." ([source](https://news.ycombinator.com/item?id=50158450))
- Outages and weak uptime (3 comments): "it's quite telling that when adding API keys for Anthropic, it presents a note inbetween reading "Experiencing Issues? Check Anthropic status"" ([source](https://news.ycombinator.com/item?id=47337161)) "how are they a near trillion dollar company with models so good they are treated as weapons yet have almost no 9's of uptime" ([source](https://news.ycombinator.com/item?id=48805343)) "Do you want your company's sales and customer service to shut down every time OpenAI or Anthropic goes down?" ([source](https://news.ycombinator.com/item?id=49568552))

## Alternatives hackers recommend

- [Mistral](https://hackerapproved.com/p/mistral-large-4.md): 314 recommend
- [DeepSeek API](https://hackerapproved.com/p/deepseek-api.md): 171 recommend
- [OpenAI API](https://hackerapproved.com/p/openai-models.md): 166 recommend
- [System One Models by TypeSafe](https://hackerapproved.com/p/system-one-models.md): 37 recommend
- [Claude API](https://hackerapproved.com/p/claude-api.md): 30 recommend

## Where to get it

- [Visit the official site](https://claude.com/platform/api)

## Cite as

Hacker Approved, "Anthropic API", https://hackerapproved.com/p/anthropic-api/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).