← All Comparisons

Claude Haiku 4.5 vs Gemini 2.5 Flash

A detailed comparison of Claude Haiku 4.5 (Anthropic) and Gemini 2.5 Flash (Google) across pricing, performance, and features.

Pricing Comparison

MetricClaude Haiku 4.5Gemini 2.5 FlashDifference
Input / 1M tokens$0.80$0.15-81%
Output / 1M tokens$4.00$0.60-85%
Context window200K1M
Max output8.192K65.536K

Benchmark Comparison

BenchmarkClaude Haiku 4.5Gemini 2.5 Flash
MMLU-Pro69.4%76%
HumanEval88.1%89.5%

Capabilities

CapabilityClaude Haiku 4.5Gemini 2.5 Flash
code
reasoning
text
tool-use
vision

Claude Haiku 4.5 Strengths

  • Very fast responses
  • Cheapest Anthropic option
  • Good for classification and extraction

Claude Haiku 4.5 Weaknesses

  • Weakest reasoning in the Claude family
  • Can struggle with nuanced instructions

Gemini 2.5 Flash Strengths

  • One of the cheapest models available
  • 1M context at budget pricing
  • Free tier available

Gemini 2.5 Flash Weaknesses

  • Weaker than Flash 3 on most benchmarks
  • Output quality inconsistent on edge cases

Quick Verdict

Best value: Gemini 2.5 Flash is the more affordable option at $0.15/$0.6 per 1M tokens.

Higher benchmarks: Gemini 2.5 Flash scores higher on average across available benchmarks (82.8% avg).

Larger context: Gemini 2.5 Flash supports 1M tokens.

Choose Gemini 2.5 Flash if cost matters most. Choose Claude Haiku 4.5 if you need the best possible quality for complex tasks.

More Comparisons