Anthropic200K contextclaude-haiku-4-5

Claude Haiku 4.5

Fast, cheap, surprisingly capable — the right pick for high-volume cost-sensitive work

Claude Haiku 4.5 is Anthropic's lightweight tier — small, fast, and now genuinely capable enough to handle production classification, extraction, and short-form generation tasks. The right pick when you have lots of requests and tight per-request budgets. Surprisingly good at tool use for its size.

Pricing

Rate	List price	Anvat effective	Savings
Input	$1.00	$0.70	30%
Output	$5.00	$3.50	30%
Cache write (5 min)	$1.25	$0.88	30%
Cache read	$0.10	$0.07	30%
All prices per million tokens (MTok). List = provider direct. Anvat effective = 30% discount applied.

Pricing verified 2026-06 · See full Anvat pricing

Strengths

Cheapest Claude — 3× cheaper than Sonnet, 15× cheaper than Opus
Fast — lowest TTFT in the Anthropic lineup
Strong tool use for its size class
Excellent for high-volume parallel workloads
Pairs well with Sonnet/Opus in router patterns (Haiku-first, escalate on complexity)

Where it underperforms

Not the model for deep reasoning or complex coding
Output quality drops on tasks needing >5 step planning
Hallucinations more common than Sonnet on factual recall

Use cases this model is the right pick for

Intent classification at scale
Data extraction from semi-structured text
Content moderation and tagging
Short-form rewriting and summarisation
First-pass routing in agent stacks (cheap classifier before escalation)

Benchmarks

coding
Coding (vs Sonnet)
Capable on short snippets
agentic
Throughput
Highest in Claude lineup
agentic
Cost per request
~$0.003 typical agent turn

Benchmark numbers self-reported by provider; verify against the latest publisher documentation before quoting.

Quickstart

Same wire format as direct provider APIs — your existing SDK code keeps working. Point at api.anvat.app/v1 and use your Anvat key.

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  baseURL: "https://api.anvat.app/v1",
  authToken: process.env.ANVAT_API_KEY,
});

// Cheap intent classifier — Haiku is perfect for this
const response = await client.messages.create({
  model: "claude-haiku-4-5",
  max_tokens: 64,
  messages: [
    {
      role: "user",
      content: `Classify intent. Reply with one word: support, billing, sales, other.

Message: "I'd like to upgrade my plan to Pro"`,
    },
  ],
});

Try Claude Haiku 4.5 — 30% off list

Same model, same quality, same wire format — at the discounted Anvat effective rate. $2 free credit on signup, no card required.

Get a key →