Home › Models › Claude Haiku 5.5
Claude Haiku 5.5 API pricing
Claude Haiku 5.5 is a model from Anthropic with a 200,000-token context window. Rates below are per 1,000,000 tokens, verified November 2026.
Price list
| Token class | Rate / 1M | Notes |
|---|---|---|
| Input | $0.10 | Prompt tokens not served from cache |
| Cached input | $0.01 | 90% below base input |
| Cache write | $0.125 | Charged once when the prefix is stored |
| Output | $0.50 | 5.0× the input rate |
| Long context input | $0.50 | Above 100,000 input tokens |
| Long context output | $2.50 | Same threshold |
| Batch API | −50% | Input and output, results within 24 h |
Prompts above 100K tokens bill at 5x the short-prompt rate ($0.50/$2.50). The only Claude model with a published input-length step-up.
What Claude Haiku 5.5 costs at three real workloads
Sticker prices rarely match the invoice. These three workloads use the rates above, with cache writes charged once per job and the batch discount applied only where the provider offers it.
| Workload | Requests | Tokens | Estimated cost |
|---|---|---|---|
| Support chat | 1,000 | 4K in / 600 out | $0.70 |
| RAG with caching | 1,000 | 50K in (45K cached) / 1K out | $1.46 |
| Nightly batch job | 10,000 | 10K in / 2K out | $10 |
Caching the 45K-token prefix saves about 74% on the RAG workload versus sending it uncached every call ($5.50 → $1.46). Batch mode halves the nightly job.
Where Claude Haiku 5.5 sits in the market
Out of 45 models in this dataset, Claude Haiku 5.5 ranks #3 cheapest by input rate, #3 by output rate, and #3 on a blended 1M-input-plus-250K-output workload.
| Model | Provider | Input | Output | Blended |
|---|---|---|---|---|
| Claude Haiku 5.5 | Anthropic | $0.10 | $0.50 | $0.225 |
| GPT-6 Luna | OpenAI | $0.10 | $0.50 | $0.225 |
| Qwen 3 32B | Alibaba (Qwen) | $0.15 | $0.60 | $0.30 |
| Qwen 3 8B | Alibaba (Qwen) | $0.05 | $0.20 | $0.10 |
Blended cost = 1M input tokens + 250K output tokens at list price.
FAQ
How much does Claude Haiku 5.5 cost per 1M tokens?
$0.10 per 1M input tokens and $0.50 per 1M output tokens on the standard tier, rising to $0.50 / $2.50 above 100,000 input tokens.
Is prompt caching cheaper on Claude Haiku 5.5?
Yes. Cached input bills at $0.01 per 1M tokens, 90.0% below the $0.10 base input rate, with a one-off cache write at $0.125 per 1M.
Does Claude Haiku 5.5 support the Batch API?
Yes. Batch requests bill at roughly half the standard input and output rates: the nightly job above costs $20 at list price and about $10 in batch mode.
Keep reading
- All Anthropic models and rates
- Every model in the dataset, ranked by price
- Claude Haiku 4.5 vs Claude Haiku 5.5
- GPT-6 Luna vs Claude Haiku 5.5
- Claude Haiku 5.5 vs Gemini 3.1 Flash-Lite
- Calculate your own monthly cost with the free estimator
Run these numbers in the calculator