Claude Haiku 4.5
Anthropic · Active · available in 31 of 31 regions
Tool useVisionPrompt cachingBatch inferenceStreaming EmbeddingsFine-tuning
- Context window
- 200K
- 200,000 tokens
- Max output
- 64K
- Input / 1M
- $1.00
- us-east-1
- Output / 1M
- $5.00
What Claude Haiku 4.5 is good for
editorialClaude Haiku 4.5 offers a 200K-token context window with tool use and prompt caching at $1.00 per 1M input tokens. Callable in 31 of 31 regions.
Suits
- Long documents. 200K tokens covers long documents and multi-file context without splitting them.
- Repeated prompts. Supports prompt caching, so a re-sent prefix costs about 10× less than fresh input. On agentic loops that resend a long system prompt every step, this moves the bill more than the headline price does.
- Tool use. Can call functions, so it can drive retrieval, look things up, and take actions rather than only answering.
- Images. Accepts image input: screenshots, scanned documents, charts and UI captures.
- Bulk jobs. Batch inference is available, typically at half the on-demand rate, for work that can wait — backfills, nightly enrichment, evaluation runs.
Think twice if
- Invocation. Reachable only through a cross-region inference profile — calling the bare model ID returns a validation error. Use an ID such as au.anthropic.claude-haiku-4-5-20251001-v1:0.
This section is judgement, not data from AWS. It is composed from the capabilities, context window, price position and region coverage shown elsewhere on this page — so it stays in step with the daily snapshot rather than going stale.
Where teams typically use it
Chat and assistants
Conversational work where responsiveness matters as much as depth.
- · A customer-facing assistant where the first token needs to appear in under a second.
- · An in-product copilot that explains what the user is looking at and answers follow-ups.
- · A triage bot that qualifies an incoming request before routing it to the right team.
Summarisation and extraction
High-volume, structured output from unstructured input. Usually the cheapest workload to run well.
- · Turning a million support tickets into a fixed schema of product, severity and root cause.
- · Pulling line items, totals and dates off scanned invoices into JSON.
- · Condensing every meeting transcript into decisions taken and owners assigned.
Model & inference profile IDs
Base model ID — direct on-demand invoke
Cross-region inference profiles (5)
Regions marked Inference profile only require a profile ID rather than the base model ID — invoking the base ID there returns a validation error. Regional (us., eu.) profiles carry a 10% premium over global.
Pricing by region
USD per 1M tokens| Region | Input | Output | Cache read | Cache write | Batch in | Batch out | Source |
|---|---|---|---|---|---|---|---|
| us-east-1 US East (N. Virginia) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| us-east-2 US East (Ohio) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| us-west-1 US West (N. California) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| us-west-2 US West (Oregon) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ca-central-1 Canada (Central) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ca-west-1 Canada West (Calgary) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| mx-central-1 Mexico (Central) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| eu-west-1 EU (Ireland) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| eu-west-2 EU (London) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| eu-west-3 EU (Paris) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| eu-central-1 EU (Frankfurt) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| eu-central-2 Europe (Zurich) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| eu-north-1 EU (Stockholm) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| eu-south-1 EU (Milan) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| eu-south-2 Europe (Spain) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-east-2 Asia Pacific (Taipei) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-northeast-1 Asia Pacific (Tokyo) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-northeast-2 Asia Pacific (Seoul) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-northeast-3 Asia Pacific (Osaka) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-south-1 Asia Pacific (Mumbai) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-south-2 Asia Pacific (Hyderabad) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-southeast-1 Asia Pacific (Singapore) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-southeast-2 Asia Pacific (Sydney) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-southeast-3 Asia Pacific (Jakarta) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-southeast-4 Asia Pacific (Melbourne) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-southeast-5 Asia Pacific (Malaysia) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-southeast-6 Asia Pacific (New Zealand) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| ap-southeast-7 Asia Pacific (Thailand) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| il-central-1 Israel (Tel Aviv) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| af-south-1 Africa (Cape Town) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
| sa-east-1 South America (Sao Paulo) | $1.00 | $5.00 | $0.100 | $1.25 | $0.500 | $2.50 | curated |
Curated pricing. The AWS Price List API publishes no SKU for this model, so these figures are hand-maintained from AWS's public pricing page and verified on Sep 15, 2026. They are not machine-derived — confirm in the AWS console before committing spend.
Region availability
Inference profile only us-east-1 Profile only Inference profile only us-east-2 Profile only Inference profile only us-west-1 Profile only Inference profile only us-west-2 Profile only Inference profile only ca-central-1 Profile only Inference profile only ca-west-1 Profile only Inference profile only mx-central-1 Profile only Inference profile only eu-west-1 Profile only Inference profile only eu-west-2 Profile only Inference profile only eu-west-3 Profile only Inference profile only eu-central-1 Profile only Inference profile only eu-central-2 Profile only Inference profile only eu-north-1 Profile only Inference profile only eu-south-1 Profile only Inference profile only eu-south-2 Profile only Inference profile only ap-east-2 Profile only Inference profile only ap-northeast-1 Profile only Inference profile only ap-northeast-2 Profile only Inference profile only ap-northeast-3 Profile only Inference profile only ap-south-1 Profile only Inference profile only ap-south-2 Profile only Inference profile only ap-southeast-1 Profile only Inference profile only ap-southeast-2 Profile only Inference profile only ap-southeast-3 Profile only Inference profile only ap-southeast-4 Profile only Inference profile only ap-southeast-5 Profile only Inference profile only ap-southeast-6 Profile only Inference profile only ap-southeast-7 Profile only Inference profile only il-central-1 Profile only Inference profile only af-south-1 Profile only Inference profile only sa-east-1 Profile only
TPM / RPM are default account quotas from AWS Service Quotas where a model-specific limit is published. They are per-account defaults and adjustable on request.
Where this model actually runs
Under a strict residency constraint, Claude Haiku 4.5 can be served without the request leaving 4 of the 25 jurisdictions with a Bedrock region.
Claude Haiku 4.5 compared
Side-by-side pages against every model people weigh this one against.
Claude Haiku 4.5 vs Claude Fable 5Claude Haiku 4.5 costs 10× less per input token than Claude Fable 5, and Claude Fable 5 holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs Claude Fable 5.1Claude Haiku 4.5 costs 10× less per input token than Claude Fable 5.1, and Claude Fable 5.1 holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs Claude Opus 4.5Claude Haiku 4.5 costs 5.0× less per input token than Claude Opus 4.5.Claude Haiku 4.5 vs Claude Opus 4.6Claude Haiku 4.5 costs 5.0× less per input token than Claude Opus 4.6, and Claude Opus 4.6 holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs Claude Opus 4.7Claude Haiku 4.5 costs 5.0× less per input token than Claude Opus 4.7, and Claude Opus 4.7 holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs Claude Opus 4.8Claude Haiku 4.5 costs 5.0× less per input token than Claude Opus 4.8, and Claude Opus 4.8 holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs Claude Opus 5Claude Haiku 4.5 costs 5.0× less per input token than Claude Opus 5, and Claude Opus 5 holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs Claude Sonnet 4.5Claude Haiku 4.5 costs 3.0× less per input token than Claude Sonnet 4.5.Claude Haiku 4.5 vs Claude Sonnet 4.6Claude Haiku 4.5 costs 3.0× less per input token than Claude Sonnet 4.6, and Claude Sonnet 4.6 holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs Claude Sonnet 5Claude Haiku 4.5 costs 2.0× less per input token than Claude Sonnet 5, and Claude Sonnet 5 holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs GPT-6 AstraClaude Haiku 4.5 costs 10× less per input token than GPT-6 Astra, and GPT-6 Astra holds 1.1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs GPT-5.6 LunaGPT-5.6 Luna costs 5.0× less per input token than Claude Haiku 4.5, and GPT-5.6 Luna holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs GPT-5.6 SolClaude Haiku 4.5 costs 4.0× less per input token than GPT-5.6 Sol, and GPT-5.6 Sol holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs GPT-5.6 TerraClaude Haiku 4.5 costs 2.0× less per input token than GPT-5.6 Terra, and GPT-5.6 Terra holds 1M tokens against 200K for Claude Haiku 4.5.Claude Haiku 4.5 vs Grok 4.6Claude Haiku 4.5 costs 2.0× less per input token than Grok 4.6, and Grok 4.6 holds 500K tokens against 200K for Claude Haiku 4.5.
Claude Haiku 4.5 appears in
Common questions
- How much does Claude Haiku 4.5 cost on Amazon Bedrock?
- $1.00 per 1M input tokens and $5.00 per 1M output tokens in us-east-1. Cached input reads at $0.100, roughly a tenth of the input rate, which dominates the bill on agentic workloads that re-send a long prefix.
- Which AWS regions support Claude Haiku 4.5?
- 31 of 31 regions: us-east-1, us-east-2, us-west-1, us-west-2, ca-central-1, ca-west-1, mx-central-1, eu-west-1, eu-west-2, eu-west-3, eu-central-1, eu-central-2, eu-north-1, eu-south-1, eu-south-2, ap-east-2, ap-northeast-1, ap-northeast-2, ap-northeast-3, ap-south-1, ap-south-2, ap-southeast-1, ap-southeast-2, ap-southeast-3, ap-southeast-4, ap-southeast-5, ap-southeast-6, ap-southeast-7, il-central-1, af-south-1, sa-east-1. Regions marked profile-only need a cross-region inference profile ID such as au.anthropic.claude-haiku-4-5-20251001-v1:0 rather than the bare model ID.
- What is the context window of Claude Haiku 4.5?
- 200,000 tokens (200K), with up to 64K output tokens per request.
- What is the model ID for Claude Haiku 4.5 on Bedrock?
- anthropic.claude-haiku-4-5-20251001-v1:0. In regions where it is only reachable through a cross-region inference profile, use a prefixed ID instead: au.anthropic.claude-haiku-4-5-20251001-v1:0, eu.anthropic.claude-haiku-4-5-20251001-v1:0, global.anthropic.claude-haiku-4-5-20251001-v1:0.