This model is deprecated. We still benchmark its performance, but less often: every 18 hours instead of every 6.

Anthropic has launched a newer release, Claude Haiku 5.5. We suggest considering it instead.

•Claude 4.5 Haiku•

Proprietary model

•

Released October 2025

Claude 4.5 Haiku (Reasoning) API Provider Benchmarking & Analysis

Model Comparison

This analysis is intended to support you in choosing the best API provider of Claude 4.5 Haiku (Reasoning) for your use-case.

Lowest Time per Task

#1
AmazonAmazon
2.2 min
#2
Google VertexGoogle Vertex
3.5 min
#3
AzureAzure
3.6 min
#4
AnthropicAnthropic
3.7 min

Time per Intelligence Index task (minutes)

Total 4 providers

Lowest Cost per Task

#1
AmazonAmazon
$0.49
#2
AzureAzure
$0.50
#3
AnthropicAnthropic
$0.53
#4
Google VertexGoogle Vertex
$0.57

Cost per Intelligence Index task

Total 4 providers

Claude 4.5 Haiku is available through 4 API providers, each offering different performance characteristics and pricing. Below is a comparison of the key metrics across providers.

  • For time per task, Amazon (2.2 min), Google Vertex (3.5 min), and Azure (3.6 min) finish an Intelligence Index task fastest.
  • For cost per task, Amazon ($0.49), Azure ($0.50), and Anthropic ($0.53) are the cheapest per Intelligence Index task.
  • Amazon is both the quickest and the cheapest per task.
Inference time (minutes) per Intelligence Index task · Lower is better
Weighted average cost (USD) per Intelligence Index task · Lower is better

Pricing

Cost per Intelligence Index Task: Claude 4.5 Haiku (Reasoning)

Weighted average cost (USD) per Intelligence Index task · Segmented by token type · Lower is better

Pricing: Cache Hit, Input, and Output: Claude 4.5 Haiku (Reasoning)

Price (USD per M Tokens) · Lower is better

Output Speed vs. Cost per Task: Claude 4.5 Haiku (Reasoning)

Output speed: output tokens divided by total request time, including time to first token · Weighted average cost (USD) per Intelligence Index task
Most attractive quadrant

Speed

Average of the single-request and multi-turn agentic workloads measured

Time per Intelligence Index Task: Claude 4.5 Haiku (Reasoning)

Output tokens divided by end-to-end output speed (minutes per Intelligence Index task) · Lower is better

Latency

Average of the single-request and multi-turn agentic workloads measured

Time to First Answer Token: Claude 4.5 Haiku (Reasoning)

Seconds until the first answer token after any reasoning · Average of the single-request and multi-turn agentic workloads measured · Lower is better

Reliability

Share of requests the endpoint failed, across all the single-request and multi-turn agentic workloads measured

Endpoint Error Rate: Claude 4.5 Haiku (Reasoning)

Share of requests the endpoint failed with an error, a rate limit or a timeout · Across all the single-request and multi-turn agentic workloads measured · Lower is better

Caching

Cache Hit Rate: Claude 4.5 Haiku (Reasoning)

Median across multi-turn agentic session runs in the last four weeks, updated Oct 26, 2026 · Higher is better

Cost per Task vs. Cache Hit Rate: Claude 4.5 Haiku (Reasoning)

Weighted average cost (USD) per Intelligence Index task · Cache hit rate: median of the last four weeks, updated Oct 26, 2026
Most attractive quadrant

Key Comparison Metrics & API Features

Amazon Bedrock logoAmazon Bedrock
$0.49
2.3
22.25
0.0%
200k
Proprietary
Anthropic logoAnthropic
$0.53
3.9
26.95
0.0%
200k
Proprietary
Google logoGoogle
$0.57
3.7
16.28
0.0%
200k
Proprietary
Microsoft Azure logoMicrosoft Azure
$0.50
3.8
30.93
0.0%
200k
Proprietary

Frequently Asked Questions

Common questions about Claude 4.5 Haiku (Reasoning) providers