Claude Sonnet 5 vs Gemini 3.7 Flash

Gemini 3.7 Flash is the cheaper of the two across every workload shape tested below. Headline rates are $2 in / $10 out for Claude Sonnet 5 versus $0.75 in / $3.75 out for Gemini 3.7 Flash, per million tokens. Note that sticker price understates the gap: Claude Sonnet 5 tokenizes the same text into roughly 30% more tokens, which the figures below account for.

Real monthly cost, three workloads

Workload Claude Sonnet 5 Gemini 3.7 Flash Cheaper
Support chatbot
1.5k in / 400 out · 100k req/mo · 30% cached
 $804.70 $232.13 Gemini 3.7 Flash (3.5× cheaper)
RAG / doc Q&A
12k in / 700 out · 50k req/mo · 60% cached
 $1,173 $338.25 Gemini 3.7 Flash (3.5× cheaper)
Coding agent
25k in / 2.5k out · 20k req/mo · 80% cached
 $1,014 $292.50 Gemini 3.7 Flash (3.5× cheaper)

Costs include prompt caching at the stated hit rate and per-model tokenizer correction. Adjust for your own traffic →

Specifications side by side

Claude Sonnet 5 Gemini 3.7 Flash
Provider Anthropic Google
Input / M tokens $2 $0.75
Output / M tokens $10 $3.75
Cached input $0.2 $0.075
Context window 1000K —
Batch discount −50% −50%
Tokenizer factor ×1.3 ×1
Min prefix to cache 1024 —
Announced retirement 2027-06-30+ —

Which should you pick?

On cost alone, Gemini 3.7 Flash wins every workload shape above.

Caching behaviour differs. Claude Sonnet 5 requires a prefix of at least 1,024 tokens before caching activates. Below those thresholds cache_control is ignored silently, with no error — which can quietly erase the saving you were counting on.

Lifecycle matters as much as price here. Claude Sonnet 5 has an announced shutdown date of 2027-06-30 or later. A model that is marginally cheaper but retires within your planning horizon is rarely the better choice.

Full detail: Claude Sonnet 5 · Gemini 3.7 Flash

All figures verified 2026-08-18 against official provider documentation. Fields a provider does not publish are shown as “—” rather than estimated.