A rate per million tokens is not a budget. The calculator takes your request volume, prompt size and cache hit rate and gives you a monthly figure.
Claude Haiku 3.5 API pricing (September 2026)
Retired 19 Feb 2026, shown at its final price for historical comparison only.
What it costs at real volumes
A chat-shaped workload: 800 input tokens and 300 output tokens per request, no caching, no batch.
Specifications
- Provider
- Anthropic Claude
- Context window
- 200K tokens
- Max output
- 8.2K tokens
- Tier
- Small
- Status
- Retired
- Blended $/1M (3:1)
- $1.60
- Cache break-even, caching each turn
- 22%
- Prices verified
- 11 Sep 2026
Source: Anthropic Claude pricing. Finitizer is not affiliated with Anthropic Claude.
Similar models
Other Claude models
- Claude Fable 5.1+$18.40
- Claude Opus 5.5+$6.40
- Claude Sonnet 5.5+$2.40
- Claude Haiku 5.5-$1.40
- Claude Opus 5+$8.40
- Claude Sonnet 5+$2.40
- Claude Opus 4.1+$28.40
- Claude Sonnet 4.5+$4.40
- Claude Haiku 4.5+$0.400
Closest prices elsewhere
- Gemini 3.8 Flash-$0.100
- Amazon Nova Pro-$0.200
- GPT-6.1 Terra+$0.400
- Mistral Large 4-$0.550
Frequently asked questions
How much did Claude Haiku 3.5 cost per 1M tokens?
Claude Haiku 3.5 cost $0.800 per million input tokens and $4.00 per million output tokens at Anthropic Claude final list prices. Output was 5.0x the input rate, so the input:output ratio of a workload drives the real cost more than the headline price.
Does Claude Haiku 3.5 support prompt caching?
Yes. Cached input is billed at $0.080 per million tokens, 10x cheaper than uncached input. Cache writes carry a premium, so caching that writes every turn (chat history, agent steps) only pays above roughly a 22% hit rate. A shared prompt reused within a few minutes pays at almost any hit rate.
What is the batch discount for Claude Haiku 3.5?
Requests submitted through the asynchronous batch endpoint are billed at 50% off both input and output rates, bringing input to $0.400 and output to $2.00 per million tokens.
What is Claude Haiku 3.5's context window?
200K tokens of context with up to 8.2K output tokens per request. A larger context window raises the ceiling on what you can send, not the price per token, but filling it on every request is one of the most common causes of an unexpected bill.
Running Claude Haiku 3.5 in production?
Finitizer Tokenomics reads what this model actually costs you from your own bills and exports, shows which workload, team or API key is spending it, flags a spike or a new spend source the day your data shows it, and forecasts where the month will land.
See Finitizer Tokenomics