Gemini 2 Pro
ActiveGoogle's flagship multimodal model with very long context.
Overview
Gemini 2 Pro is Google's top general-purpose model with up to 2M context tokens, multimodal inputs, tool calling, and structured outputs.
Benchmarks
| Benchmark | Score | Source |
|---|---|---|
| GPQA DiamondReasoning | 68.1% accuracy | Self-reported Google model card |
| GSM8KMath | 94.4% accuracy | Self-reported Google model card |
| HumanEvalCoding | 88.4pass@1 % | Self-reported Google model card |
| MMLUGeneral knowledge | 85.2% accuracy | Third-party Artificial Analysis |
| MMLU-ProGeneral knowledge | 75.8% accuracy | Self-reported Google model card |
Integrations & tooling support
- Tool calling
- Supported
- Structured outputs
- Supported
Price vs quality
Poor value
Below-average benchmark performance for the price.
- Quality percentile
- 21.5%
- Effective price
- $4.0625/1M
- Pricing breakdown
- $1.25/1M in
$5/1M out
vs 5 benchmarks
/ 1M tokens (input + 3× output)
Community ratings
No ratings yet. Be the first to rate Gemini 2 Pro.
Rate Gemini 2 Pro
Sign in to rate and review.
Comments
Sign in to leave a comment.