Anthropic
Updated May 2026Claude
The safety-first frontier family
Anthropic was founded in 2021 by former OpenAI leaders who wanted a more cautious, research-driven approach to building powerful AI. Their training method, Constitutional AI, teaches the model to follow a written set of principles instead of relying only on human feedback. The result is a family known for a measured tone, careful reasoning, strong writing, and an unusually long attention span on complex documents.
Variants
4
Best at
long-document reasoning
Max context
1M
tokens
Pricing
Free
$0
per limited daily messages at claude.ai
Pro
$20
per month
Max
$100-$200
per month for power users
Opus API
$5 in / $25 out
per million tokens
Sonnet API
$3 in / $15 out
per million tokens
Haiku API
$1 in / $5 out
per million tokens
Variants
Sort the table by context window or cost to find the right variant. Click any version below for a battle card with ranks, pricing notes, and official links.
| Modalities | ||||
|---|---|---|---|---|
Claude Opus 4.7 claude-opus-4-7 | 1M | $5.00 / $25 | 2026 | textvisioncode |
Claude Sonnet 4.6 claude-sonnet-4-6 | 1M | $3.00 / $15 | 2025 | textvisioncode |
Claude Haiku 4.5 claude-haiku-4-5 | 200K | $1.00 / $5.00 | 2025 | textvision |
Claude 3.5 Sonnet claude-3-5-sonnet | 200K | varies / varies | 2024 | textvision |
Battle card
Context rank
#1
within Claude
Capability rank
#1
modalities + reasoning
Weights
Closed
API/app access
Best fights to pick
- drafting legal briefs
- multi-file code refactors
- analyzing 500-page research reports
- building agentic coding assistants
Rankings are Tendril directory ranks, computed from the model data shown here. Public benchmark leaderboards change often, so official docs and current benchmark pages should be checked before buying or deploying.
Learn
Lessons about this model
Structured lessons that cover Claude directly or put it in context alongside its rivals.
Claude Opus 4.7 — when extended thinking earns its cost
Claude Opus 4.7 vs. Sonnet 4.6 — which Claude to pick
Claude Haiku 4.5 — speed/cost analysis
Claude Opus 4.7 — extended thinking cost math
Claude Artifacts — when AI builds alongside you
Claude Skills — reusable specialized agents
GPT-5.5 vs. Claude Opus 4.7 — which chatbot wins your day
Anthropic Prompting Patterns
Check yourself
Quizzes
Short, mixed-difficulty quiz sets on Claude and its model family.
Anthropic's Claude Family
6 questions
From Claude 1 to the current generation — Haiku, Sonnet, Opus.
Start quiz →Hands-on
Try these prompts
Ready-made prompts that show Claude at its best. Use them in your own AI workspace, then compare the output with what you learned in Tendril.
Extended thinking on a tricky logic puzzle
CreatorsWatch Claude Opus work through a multi-step constraint problem with visible reasoning before answering.
I need to schedule 5 meetings today. Meeting A must be before B. C can't happen between 10am and noon. D must be at least 90 minutes after A. E is 30 minutes and must be the last meeting of the day. All meetings are 60 minutes unless noted. Work day is 9am-5pm. Give me one valid schedule and explain any constraints that are tight.
Ask Claude to critique its own draft
BuildersUse Claude's tendency toward honest self-review to catch weak writing before it ships.
Here's a cold email draft. First, write three specific, named critiques of it — things that would make a reader roll their eyes or stop reading. Then rewrite it applying those critiques. Draft: 'Hi! I came across your company and was really impressed. I'd love to hop on a quick call to learn more about your needs. Are you free sometime this week?'
Claude as a code reviewer
BuildersPaste a short function and ask Claude for a diff-style review — it's particularly good at spotting edge cases politely.
Review this function as if you're a senior engineer on my team. Point out bugs, edge cases I missed, and places the code is harder to read than it needs to be. Be direct but kind.
function avgAge(users) {
let total = 0;
for (let i = 0; i < users.length; i++) {
total += users[i].age;
}
return total / users.length;
}Haiku vs. Opus on the same task
ExplorersCompare side-by-side how Haiku 4.5 and Opus 4.7 handle the same request — feel the speed and quality tradeoff yourself.
Explain what a mortgage is to a 14-year-old in under 150 words. Use one concrete example with round numbers.
Print & keep
Printable reference
One-page summaries and flowcharts — great for desks, classrooms, or study sessions.
Go deeper
Official resources
Straight from the lab — docs, API references, and the chat surfaces you can try today.
Strengths
- long-document reasoning
- writing quality
- code generation with agentic tool-use
Limits
- higher per-token cost than Chinese open models
- no image generation
- no native voice
