Gemini 2.5 Flash vs Gemini 3 Flash preview
Gemini 2.5 Flash vs Gemini 3 Flash preview: Gemini 2.5 Flash is cheaper by 40% on average. Gemini 2.5 Flash from Google Vertex AI (1,048,576-token context, reasoning, tool calls) vs. Gemini 3 Flash preview from Google Vertex AI (1,048,576-token context, reasoning, tool calls). Use Agent Command Center to A/B both in shadow mode and pick the winner per workload.
Side-by-side cost
Live workload comparison
Same workload run through both models. The cheaper one is highlighted.
3,000
01,048,576
400
065,535
5,000
01,000,000
At this workload, Gemini 2.5 Flash is 30% cheaper than Gemini 3 Flash preview — a savings of $122/month ($1,461/year).
Production recipe — Agent Command Center
strategy: cost-optimized
primary:
model: gemini-2-5-flash
provider: vertex-ai
fallback:
model: gemini-3-flash-preview
provider: vertex-ai
shadow: { sample_rate: 0.05 } # mirror 5% of traffic to compare quality live| Gemini 2.5 Flash | Gemini 3 Flash preview | |
|---|---|---|
| Input price | $0.300/M | $0.500/M |
| Output price | $2.50/M | $3.00/M |
| Context window | 1,048,576 | 1,048,576 |
| Max output | 65,535 | 65,535 |
| Function calling | ✓ | ✓ |
| Vision | ✓ | ✓ |
| Audio input | — | ✓ |
| Reasoning | ✓ | ✓ |
| Prompt caching | ✓ | ✓ |
| Structured output | ✓ | ✓ |
| Pricing verified | May 12, 2026 | May 12, 2026 |
Benchmark comparison
Side-by-side public benchmark scores. Greener bar = winner.