LLMCostLab

PricesCompare → Claude Haiku 4.5 vs Gemini 2.5 Flash-Lite

Claude Haiku 4.5 vs Gemini 2.5 Flash-Lite: API cost compared

Claude Haiku 4.5 lists at $1.000 in / $5.000 out per 1M tokens. Gemini 2.5 Flash-Lite lists at $0.100 in / $0.400. Sticker price only tells you so much — below is what each actually costs across eight production workloads.

advertisement
Cheaper on more workloads
Gemini 2.5 Flash-Lite
wins 8 of 8
Blended rate gap
11.4x
at 3:1 in:out

Side by side

Claude Haiku 4.5Gemini 2.5 Flash-Lite
ProviderAnthropicGoogle
Input per 1M$1.000$0.100
Output per 1M$5.000$0.400
Cached input per 1M$0.100not published
Context window200k1M
Blended per 1M (3:1)$2.000$0.175

Monthly cost by workload

Caching applied at each workload's typical hit rate. This is where the two models actually separate.

WorkloadClaude Haiku 4.5Gemini 2.5 Flash-LiteCheaperDifference
Customer support chatbot$128.90$16.00Gemini 2.5 Flash-Lite$112.90 (88%)
RAG document Q&A$184.00$20.80Gemini 2.5 Flash-Lite$163.20 (89%)
Coding agent$92.50$14.80Gemini 2.5 Flash-Lite$77.70 (84%)
Bulk summarization$1,346$136.00Gemini 2.5 Flash-Lite$1,210 (90%)
Structured data extraction$1,745$225.00Gemini 2.5 Flash-Lite$1,520 (87%)
Long-form content generation$214.60$18.00Gemini 2.5 Flash-Lite$196.60 (92%)
Email triage agent$402.50$62.00Gemini 2.5 Flash-Lite$340.50 (85%)
Translation pipeline$1,405$118.50Gemini 2.5 Flash-Lite$1,286 (92%)
advertisement

How to read this

Gemini 2.5 Flash-Lite is cheaper on the majority of these workloads, but "majority" is not the decision. The output:input price ratio is 5.0 for Claude Haiku 4.5 and 4.0 for Gemini 2.5 Flash-Lite. If your traffic is output-heavy — long-form drafting, translation — the model with the lower output price wins regardless of what the input price suggests. If you are input-heavy — RAG, bulk summarization, agents re-reading a codebase — the cached input rate matters more than either headline number.

Cost is also not the only axis. This site does not benchmark quality, latency, or rate limits, and a model that is 3x cheaper but needs two attempts per task is not cheaper. Use these numbers to size a bill, then validate on your own evals.

Run your own numbers

Cost calculator

Enter your own numbers, or start from a workload preset. Costs are monthly and update as you type.

Cheapest model
Cheapest → priciest spread
ModelIn /1MOut /1MMonthlyvs best