Skip to content

Claude 3.5 Haiku

budget

Anthropic•Released on 2024-11-04

Anthropic's fastest and most affordable model. Claude 3.5 Haiku matches Claude 3 Opus on many benchmarks while being significantly cheaper and faster. Ideal for high-volume tasks, quick responses, and cost-sensitive applications. Best-in-class speed-to-performance ratio.

77
Overall Score

Core Specs

200K
Context Window
8K
Max Output
✗
Reasoning
✗
Open Source
Multimodal Support
textimage

Scenario Scores

Coding
#24 / 2878
Writing
#23 / 2772
Translation
#20 / 2780
Data Analysis
#25 / 2874
Conversation
#21 / 2780
Image Gen
—

User Feedback Highlights

Based on community feedback. Hover to see original reviews.

+ Great for high-volume production− Not ideal for multi-step agentic tasks+ Fast response times+ Extremely affordable ($0.80/$4 per MTok)+ Matches Opus 3 on many tasks− Less capable than Sonnet 4.x for complex reasoning+ 200K context window− Smaller output limit (8K)
Sentiment:👍 70%😐 22%👎 8%

Pros & Cons

Pros

  • +Extremely affordable ($0.80/$4 per MTok)
  • +Fast response times
  • +200K context window
  • +Matches Opus 3 on many tasks
  • +Great for high-volume production

Cons

  • −Less capable than Sonnet 4.x for complex reasoning
  • −Smaller output limit (8K)
  • −Not ideal for multi-step agentic tasks

Reliability

SLA99.9%
Incidents (30d)1
与其他 Claude 模型共享基础设施
View Status Page →

Pricing

Input (per 1M tokens)$0.80
Output (per 1M tokens)$4.00
Subscription$20/month
Free trial available
Updated on 2026-03-08

Get Started

1Visit the provider's website
2Create an account
3Start using the model

Compare with Others

Claude 3.5 Haiku vs GPT-5.4 →Claude 3.5 Haiku vs GPT-5.4 Thinking →Claude 3.5 Haiku vs Claude Opus 4.6 →

Benchmarks

mmlu84.2%
humanEval88.1%