Models GPUs Systems AI Setups Builds Methodology

Loading developer profiles...

Loading content...

Resources

Methodology
Submit Benchmark
About

Browse

AI Models
GPUs
PC Builds

Community

Leaderboard

Legal

Privacy
Terms
Contact

We earn from qualifying purchases through affiliate links at no extra cost to you. This supports our free content and research.

Loading developer profile...

Loading developer profiles...

Loading developer profile...

google AI Models & GPU Requirements - localai.computer

Home
Home
Developers
google

google

google publishes models from 0B to 27B parameters focused on lightweight edge deployment.

7 models0B to 27B parametersPricing soon

Models

Search and sort every tracked release before opening a dedicated requirements page.

google/embeddinggemma-300m

Minimum VRAM: Q4 · 1GB

Throughput: 970.7 tok/s

From $329View requirements →

google/gemma-2-9b-it

Minimum VRAM: Q4 · 5GB

Throughput: 603.7 tok/s

From $329View requirements →

google/gemma-2-27b-it

27B

Minimum VRAM: Q4 · 14GB

Throughput: 420.2 tok/s

From $329View requirements →

google/gemma-3-270m-it

Minimum VRAM: Q4 · 4GB

Throughput: 739.8 tok/s

From $329View requirements →

google/gemma-2b

Minimum VRAM: Q4 · 1GB

Throughput: 964.5 tok/s

From $329View requirements →

google/gemma-2-2b-it

Minimum VRAM: Q4 · 1GB

Throughput: 861.7 tok/s

From $329View requirements →

google/gemma-3-1b-it

Minimum VRAM: Q4 · 1GB

Throughput: 952.3 tok/s

From $329View requirements →

Recommended hardware

Budget$300-$700

AMD Instinct MI300X • Pricing soon

1GB VRAM target • Q4

Edge inference, personal copilots, prototype agents.

Mid-range$700-$2,000

AMD Instinct MI300X • Pricing soon

2GB VRAM target • Q8

Team chatbots, offline knowledge bases, enterprise POCs.

High-end$2,000+

AMD Instinct MI300X • Pricing soon

4GB VRAM target • FP16

Agent orchestration, evaluation pipelines, long-context tasks.