Back to Models
google logo
google/gemma-3-27b-it
Not Available

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, including structured outputs and function calling. Gemma 3 27B is Google's latest open source model, successor to [Gemma 2](google/gemma-2-27b-it)

3/12/2025
128,000 tokens
#95 Vision (Overall)

Specifications

Modalities

Input
text
image
Output
text

Supported Parameters

frequency_penalty
logit_bias
max_tokens
min_p
presence_penalty
repetition_penalty
response_format
seed
stop
structured_outputs
temperature
tool_choice
tools
top_k
top_p

Max Output Tokens

65,536

Frequently asked questions

Is Gemma 3 27B available on FastMetal?
Not at the moment. GLM 5.3 Flash, from the same lab, is available on the FastMetal API today.
What is the context window of Gemma 3 27B?
128,000 tokens, shared between the prompt and the response.
How does Gemma 3 27B rank?
#148 on the public arena's Japanese board (ELO 1,309). Ranks move as the leaderboard is updated.

Leaderboard

Text
OverallELO: 1,365
#186
JapaneseELO: 1,309
#148
ChineseELO: 1,371
#200
KoreanELO: 1,323
#150
EnglishELO: 1,376
#193
frenchELO: 1,386
#164
germanELO: 1,369
#147
spanishELO: 1,357
#177
russianELO: 1,366
#173
CodingELO: 1,358
#245
MathELO: 1,322
#211
Creative WritingELO: 1,348
#166
Instruction FollowingELO: 1,343
#198
Hard PromptsELO: 1,366
#203
Multi-TurnELO: 1,358
#188
Vision
OverallELO: 1,159
#95