---
id: "google/gemma-3-27b-it"
also: ["google/gemma-3-27b-it", "hf/google/gemma-3-27b-it", "requesty/google/gemma-3-27b-it"]
cheapest_list_usd_per_m: "0.08 in, 0.16 out, at deepinfra on hf"
cheapest_effective_usd_per_m: "0.0624 in, 0.4483 out, at parasail/fp8 on openrouter"
context: 131072
max_output: 117964
input: ["text", "image"]
output: ["text"]
released: "2025-03-12"
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "logit_bias", "logprobs", "max_tokens", "min_p", "presence_penalty", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 9
gateways: "openrouter (6), hf (2), requesty (1)"
no_training: 7
trains: 0
retains_prompts: 0
coding: 10.1000
agentic: 0.1000
---
# Google: Gemma 3 27B

Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities,...