Major Release Google · Released
Gemini 1.5 Pro
What is Gemini 1.5 Pro?
Gemini 1.5 Pro is a Major Release AI model from Google, released on February 15, 2024. It has a 1M tokens context window, and scores 81.9% on MMLU. It is available via the company's API.
Google's first 1M-context model — multimodal needle-in-haystack champion
Google's next-generation model with breakthrough long context capabilities
Key features
- 1M token context window
- Multimodal understanding
- Video analysis
- Audio processing
How much does Gemini 1.5 Pro cost?
| Metric | Price |
|---|---|
| input (≤128K) | $1.25/M |
| input (>128K) | $2.50/M |
| output | $5.00/M |
Pricing data verified .
What context window does Gemini 1.5 Pro have?
- Context window
- 1M tokens
- Max output
- 8192 tokens
- Input modalities
- text, image, audio, video
- Output modalities
- text
How does Gemini 1.5 Pro score on benchmarks?
| Benchmark | Score |
|---|---|
| MMLU | 81.9% |
| HumanEval | 71.9% |
| MATH | 58.5% |
Benchmark scores verified .
What's new
- 1M token context
- Multi-hour video understanding
- MoE architecture
How does Gemini 1.5 Pro compare to Gemini Pro?
- HumanEval: 67.7% → 71.9% (+4.2pt)
- Context window: 32K tokens → 1M tokens