google Provider Benchmarks

Comprehensive performance summary covering 7 models.

This provider hub highlights throughput and latency trends across every google model monitored by LLM Benchmarks. Use it to compare hosting tiers, track regressions, and discover the fastest variants in the catalogue.

Visit google Official Website

Provider Snapshot

Models Tracked

7

Avg Tokens / Second

57.23

Avg Time to First Token (ms)

831.43

Last Updated

Feb 8, 2026

Key Takeaways

  • 7 google models are actively benchmarked with 839 total measurements across 828 benchmark runs.

  • undefined leads the fleet with 73.70 tokens/second, while gemini-2.5-pro delivers 41.00 tok/s.

  • Performance varies by 79.8% across the google model lineup, indicating diverse optimization strategies for different use cases.

  • Avg time to first token across the fleet is 831.43 ms, showing good responsiveness for interactive applications.

  • The google model fleet shows consistent performance characteristics (22.9% variation coefficient), indicating standardized infrastructure.

Fastest Models

ProviderModelAvg Toks/SecMinMaxAvg TTF (ms)
google73.7034.90105.00590.00
googleclaude-3-haiku67.6027.8078.30530.00
googlegemini-2.0-flash61.408.1487.70590.00
googlegemini-2.0-flash-lite61.1011.7080.30610.00
googlegemini-2.5-flash61.107.9578.001190.00
googlegemini-2.5-pro41.0015.7062.101590.00

All Models

Complete list of all google models tracked in the benchmark system. Click any model name to view detailed performance data.

ProviderModelAvg Toks/SecMinMaxAvg TTF (ms)
googlegemini-2.0-flash-lite61.1011.7080.30610.00
googlegemini-2.0-flash61.408.1487.70590.00
googleclaude-3-5-sonnet34.7024.4045.70720.00
googleclaude-3-haiku67.6027.8078.30530.00
google73.7034.90105.00590.00
googlegemini-2.5-flash61.107.9578.001190.00
googlegemini-2.5-pro41.0015.7062.101590.00

Featured Models

Frequently Asked Questions

Based on recent tests, undefined shows the highest average throughput among tracked google models.

This provider summary aggregates 839 individual prompts measured across 828 monitoring runs over the past month.