PRICING SNAPSHOT 2026-07-11CLAUDE FABLE 5 $10.00/$50.00 MTOKCLAUDE OPUS 4.8 $5.00/$25.00 MTOKCLAUDE SONNET 5 $2.00/$10.00 MTOKCLAUDE HAIKU 4.5 $1.00/$5.00 MTOKGPT-5.6 $5.00/$30.00 MTOKGPT-5.5 $5.00/$30.00 MTOKGPT-5.4 $2.50/$15.00 MTOKGEMINI 3.1 PRO $2.00/$12.00 MTOKGEMINI 3.5 FLASH $1.50/$9.00 MTOKGEMINI 3 FLASH $0.50/$3.00 MTOKDEEPSEEK V4 PRO $0.43/$0.87 MTOKDEEPSEEK V4 FLASH $0.14/$0.28 MTOK
TOKENBURN_INDEX

THE TICKER · ANTHROPIC · CHECKED DAILY · LAST CHANGE 2026-07-11

CLAUDE HAIKU 4.5 API PRICING

What Anthropic charges for Claude Haiku 4.5, in USD, verified daily against public list prices. Effective price — if there's a promo running, this is the promo.

INPUT / 1M TOKENS
$1.00
OUTPUT / 1M TOKENS
$5.00
INPUT / 1K TOKENS
$0.0010
OUTPUT MULTIPLE
5.0x
output vs. input price

WHAT THINGS ACTUALLY COST ON CLAUDE HAIKU 4.5

ScenarioTokens in / out1 run1,000 runs
A typical prompt (300 tokens in)300 / ~900$0.0048$4.80
A 20-turn conversation (context re-sent)4,000 / ~4,000$0.02$24.00
Summarizing a 10-page document8,000 / ~3,200$0.02$24.00
One agent run (tools, retries, thinking)2,000 / ~12,000$0.06$62.00
A code review request1,500 / ~3,750$0.02$20.25

Output sizes assume the same editorial ratios as the Burn Calculator (3x for prompts, 6x for agent runs). Estimates, not invoices.

VS. THE REST OF THE BOARD

ModelInput / MTokOutput / MTokSame job costs
Claude Fable 5 Anthropic$10.00$50.0010.0x more
Claude Opus 4.8 Anthropic$5.00$25.005.0x more
GPT-5.6 OpenAI$5.00$30.005.9x more
GPT-5.5 OpenAI$5.00$30.005.9x more
GPT-5.4 OpenAI$2.50$15.003.0x more
Claude Sonnet 5 Anthropic$2.00$10.002.0x more
Gemini 3.1 Pro Google$2.00$12.002.4x more
Gemini 3.5 Flash Google$1.50$9.001.8x more
Gemini 3 Flash Google$0.500$3.001.7x less
DeepSeek V4 Pro DeepSeek$0.435$0.8705.3x less
DeepSeek V4 Flash DeepSeek$0.140$0.28016.3x less

“Same job” = a 1,000-token prompt with a 3,000-token reply, priced on both models. Cheaper is not always better. It is, however, always cheaper.

NOW FIND OUT HOW MUCH OF THAT IS WASTE

Claude Haiku 4.5's price is fixed. Your token count isn't. Paste a real prompt into the Burn Calculator and see what it costs on Claude Haiku 4.5 — and how much a leaner version would save at 1,000 runs. Or pull the whole board from the ticker.