Google Gemini 2.5 Pro

Google Gemini 2.5 Pro: $1.25/1M input tokens, 380 tok/s throughput, 800ms latency. Compare on BenchNode.io.

$1.25

per 1M input tokens

$5 / 1M output

Specifications

model id: gemini-2.5-pro
model family: Gemini 2.5
parameters: undisclosed
context window tokens: 2,000,000
modality: text + vision + audio
reasoning: Yes

Performance

Latency (TTFT) 800 ms

lower is better

Uptime SLA 99.95%

Throughput 380 tok/s

800

ms latency

99.95%

uptime

Pricing Detail

input per 1m tokens usd: 1.25
output per 1m tokens usd: 5
free tier available: No
rate limit tpm: 4,000,000
rate limit rpm: 150
context caching input usd: 0.31

AI Analysis · gpt-4o-mini

Technical Verdict

The input cost of $1.25 / 1M tokens and output cost of $5.0 / 1M tokens position this API tier at a premium price point, while the throughput of 380 tok/s is relatively fast for real-time applications. The rate limits of 4,000,000 TPM and 150 RPM indicate suitability for high-volume production environments, with the 800 ms first-token latency favoring real-time streaming or chat applications.