Gemma 3 27B IT — AI Model Comparison

Author

Context128k

Tool calling

Google’s latest open-source multimodal model, Gemma 3 27B, supports vision-language input and text outputs, handles context windows up to 128k tokens, and understands over 140 languages. Offers improved math, reasoning, and chat capabilities, including structured outputs and function calling.

Tokens, last 14 days

Prompt56.4K

Completion7.04K

Total63.5K

Performance

Latency (p50)0.23 s

Throughput (p50)56.8 tok/s

Pricing

Input$0.04/M tokens~~market price $0.08/M tokens~~

Output$0.08/M tokens~~market price $0.16/M tokens~~

Cached input$0.02/M tokens~~market price $0.04/M tokens~~

Features

Inputtext, image

Outputtext

APIResponses, Chat Completions, Messages

Vision

Open model page