Skip to main content
GoogleNov 4, 2025

Google Gemma 3 27B Instruct

google-gemma-3-27b-itPrivateOpen weights

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, including structured outputs and function calling. Gemma 3 27B is Google’s open source model, successor to Gemma 2.

$0.12/$0.20
per 1M in / out
198K
context window
16K
max output
How it comparesamong 143 text models
GPQA Diamondhigher than 8%48.0%
Blended pricecheaper than 97%$0.14
Context windowlarger than 14%198K
Max outputlarger than 8%16K

Further right is better. Hover a dot for its model.

Time to first token— 
Output speed— 
Uptime— 

Last 24 hours of Venice traffic.How it’s measured

Capabilities

What the model takes in, returns and supports through the API.

InTextImage
OutText

Not supportedreasoning, video input, audio input, X search, prompt caching and log probabilities

Benchmarks

GPQA Diamond
48.0%128th of 139 · OpenAI GPT OSS 120B and NVIDIA Nemotron 3 Nano 30B score higher for less

Specs

Model ID
google-gemma-3-27b-it
Endpoint
POST /api/v1/chat/completions
Context window
198,000 tokens
Max output
16,384 tokens
Reasoning
None
Precision
FP8
Privacy
Private
Added
Nov 4, 2025

Compare

ModelInputOutputGPQA DiamondContextMax outputFirst tokenSpeed
Google Gemma 3 27B Instruct$0.12$0.2048.0%198K16K——
OpenAI GPT OSS 120B$0.07$0.3066.7%128K16K——
Mistral Small 3.2 24B Instruct$0.0938$0.2541.4%256K16K——
NVIDIA Nemotron 3 Nano 30B$0.075$0.3049.8%128K16K——
GLM 4.7 Flash Heretic$0.07$0.4035.4%200K24K——

Other versions

  1. Google Gemma 3 27B InstructNov 2025
  2. Google Gemma 4 31B InstructApr 2026

Pricing

Billed per token in USD. Reasoning tokens bill as output.

Input$0.12per 1M tokens
Output$0.20per 1M tokens

Estimate a workload

$0.0003per request
$3.40for 10,000 requests

Blended price against similar models

Google Gemma 3 27B Instruct$0.14

Per 1M tokens at three input tokens for each output token.

Quickstart

OpenAI-compatible: point any OpenAI SDK at the Venice base URL with a Venice API key.

curl https://api.venice.ai/api/v1/chat/completions \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "google-gemma-3-27b-it",
"messages": [{ "role": "user", "content": "Explain TEE attestation in two sentences." }]
}'
Base URL https://api.venice.ai/api/v1

Parameters

ParameterValuesNotes
modelgoogle-gemma-3-27b-itRequired
messagestext · image_urlContent parts the model reads
max_completion_tokensUp to 16,384
toolsFunction definitionsWith tool_choice
response_format{ type: "json_schema" }Output that matches a schema
streamtrue | falseServer-sent events
venice_parameters.enable_web_searchauto | on | offWeb results and citations