Local LLM model page

Gemma 2 (27B)

Google's large Gemma 2. Excellent reasoning and coding. Strong performance at 27B.

Find the best model for my hardware Browse all 183 LLMs

Parameters

27B

Minimum RAM

24 GB

Model size

16 GB

Quantization

Q5_K_M

Can Gemma 2 (27B) run locally?

Gemma 2 (27B) is best suited for power-user machines with 32 GB RAM. LocalClaw recommends Q5_K_M as the default quantization, with at least 24 GB RAM.

Search term for LM Studio or compatible runtimes: gemma-2-27b-it

Hugging Face repository: lmstudio-community/gemma-2-27B-it-GGUF

chatcodepowerquality

Strengths

Near-frontier quality at 27B
Strong reasoning and coding
Efficient architecture

Limitations

8K context limit
Needs 24GB+ RAM
Gemma license restrictions

Best use cases

Advanced reasoning
Professional coding assistance
Research
Content creation

Benchmarks

Speed: 4/10

Quality: 9/10

Coding: 8/10

Reasoning: 8/10

Technical details

Developer: Google DeepMind

License: Gemma License

Context window: 8,192 tokens

Architecture: Transformer (decoder-only) with sliding window attention

Released: 2024-06