Google DeepMind·Gemma

Gemma 2 27B

Raised reasoning quality in 9B and 27B sizes designed for accessible accelerators.

LegacyLanguageReasoningOpen weights

MODEL NUMBER102

Specifications

Only published specifications are shown.

Total parameters

27B

Architecture

Dense

Context

Input modalities

Text

Output modalities

Text

Open weights

Yes

License

Gemma Terms

Status

Legacy

Raised reasoning quality in 9B and 27B sizes designed for accessible accelerators.

Compared with Gemma ↗

PreviousGemma7B · 8K ctx

CurrentGemma 2 27B27B · 8K ctx

New capabilities Reasoning

ReasoningOn-deviceFine-tuning

2024-06-27Open weights

Gemma 2 27B was released.

Select a card to view the related model.