How These Compare to the Competition

Here's how the new Gemini budget models stack up against other cheap options:

• Gemini 2.5 Flash-Lite vs DeepSeek V4 Flash: Gemini is cheaper on input ($0.10 vs $0.14) but more expensive on output ($0.40 vs $0.28). DeepSeek wins on output-heavy tasks.
• Gemini 3.1 Flash-Lite vs Claude Haiku 4.5: Gemini is 75% cheaper on input ($0.25 vs $1.00) and 70% cheaper on output ($1.50 vs $5.00). Haiku has better reasoning, but for most tasks Gemini is the better value.
• Gemini 3 Flash vs GPT-4o mini: GPT-4o mini is cheaper at $0.15/$0.60 vs Gemini 3 Flash at $0.50/$3.00. But Gemini 3 Flash has 1M context vs GPT-4o mini's 128K — 8x more context for 3x the price.
• All 3 vs Mistral Small 4: Mistral Small 4 at $0.15/$0.60 is cheaper on input than Gemini 3.1 Flash-Lite ($0.25/$1.50) but has only 128K context vs 1M. For long documents, the new Gemini models win.

Which One Should You Use?

The Bigger Picture

These 3 new models push the APIpulse database to 87 models across 10 providers. The Gemini lineup now covers every price point from $0.10 to $12.50 per 1M input tokens — the widest range of any single provider.

The trend is clear: 1M context windows are becoming the standard, even at budget price points. If you're still using models with 128K context, you're paying for less capability than you could get.

With Claude 4 retired on June 15, now is the perfect time to evaluate these new Gemini options. Use our deprecation calculator to see how much you'd save by switching.