Loading...
About
Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy...
Related Models
Google: Gemini 3.1 Flash Lite2026-05-07Google: Gemini 3.1 Flash Lite Preview2026-03-03Google: Gemini 3.1 Pro Preview Custom Tools2026-02-25Google: Gemini 3.1 Pro Preview2026-02-19Gemini 3 Deep Think2026-02-05Gemini 3 Flash Preview (Reasoning)2025-12-17Gemini 3 Flash Preview (Non-reasoning)2025-12-17Google: Gemini 3 Flash Preview2025-12-17
Pricing
Input
$1.25
per 1M tokens
Output
$10.00
per 1M tokens
Blended
$3.44
per 1M tokens
Cheaper than 18% of models. Median price is $0.56/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$3.44
Monthly
$103.13
vs. Similar Models
Anthropic: Claude 3.7 Sonnet (thinking)Q:+0.1
$6.00+75%
Nova 2.0 Lite (high)Q:-0.1
$0.85-75%
Gemini 3 Flash Preview (Non-reasoning)Q:+0.4
$1.13-67%
GLM-4.7 (Non-reasoning)Q:-0.4
$1.00-71%
Performance
124
tokens/sec
Faster than 68% of models
17.99
seconds
Faster than 8% of models
17.99
seconds
Faster than 28% of models
Market Median
86 tok/s
44% faster
Median TTFT
1.07s
1579% slower
Throughput/Dollar
36
tok/s per $/1M
Speed Comparison
OpenAI: GPT-5 Nano
123 tok/s-0%
Meta: Llama 4 Scout
123 tok/s-0%
NVIDIA Nemotron Nano 9B V2 (Reasoning)
124 tok/s+1%
Context Window
1.0M
tokens
Larger than 91% of models
Max Output
66K
tokens
6% of context
Benchmarks
MMLU-Pro
86.2%
GPQA Diamond
84.4%
HLE
21.1%
LiveCodeBench
80.1%
SciCode
42.8%
TerminalBench Hard
26.5%
MATH-500
96.7%
AIME
88.7%
AIME 2025
87.7%
IFBench
48.7%
Long Context Recall
66.0%
Tau2
54.1%
Market AverageTop Score
