v2.2 · 2026-06-12
Gemma 2 2B Instruct
gemma-2-2b-it
Gemma 2 2B Instruct is a 2B text chat model profile for local AI planning. This page records provider, license, context, quantization, hardware-fit estimates, setup hints, source links, and caveats so readers do not choose a model by name alone.
PROVIDER
Google
FAMILY
Gemma 2
MODEL TYPE
text-chat
PARAMETERS
2B
MODALITIES
text
ARCHITECTURE
decoder-only Transformer text-to-text language model
CONTEXT WINDOW
8,192 tokens
TRAINING TOKENS
2.0T
RELEASE DATE
2024-07-31
GGUF REPOSITORIES
bartowski/gemma-2-2b-it-GGUF community conversion
Community llama.cpp GGUF quantization of the official Google instruction-tuned checkpoint.
SETUP HINTS
OLLAMA
ollama run gemma2:2bLM STUDIO
Search for bartowski/gemma-2-2b-it-GGUF and select Q4_K_M or Q5_K_M.
LLAMA.CPP
Use hf.co/bartowski/gemma-2-2b-it-GGUF:Q4_K_M; increase context only if memory permits.QUANTIZATION FILE SIZES (GGUF)
Q4_K_M
1.71 GB
Q5_K_M
1.92 GB
Q8_0
2.78 GB
Quantization note: Q4_K_M for default local use; Q5_K_M if the model still fits comfortably in RAM/VRAM.
RAM / VRAM ESTIMATES
MIN RAM
4 GB
COMFORTABLE RAM
8 GB
MIN VRAM
3 GB
COMFORTABLE VRAM
4 GB
These are conservative local-inference estimates, not official hardware requirements.
HARDWARE FIT
CPU only (no GPU)
yes
8 GB RAM
yes
16 GB RAM
yes
32 GB RAM
yes
8 GB VRAM
yes
12 GB VRAM
yes
24 GB VRAM
yes
Apple Silicon (unified memory)
yes
Hardware fit values are conservative local-inference estimates based on GGUF size plus runtime and KV-cache overhead. Actual requirements depend on context length, quantization, runtime, GPU offload, and other running apps.
BEST FOR
·fast local chat
·summarization
·drafting
·small-device experimentation
AVOID IF
·image input is required
·very long context is required
·native tool/function calling is required
CAVEATS
·Gemma 2 is text-only; do not treat it as a vision-language model.
·GGUF files are community conversions, not official Google GGUF releases.
·RAM and VRAM figures are estimates based on GGUF file size plus conservative runtime and KV-cache overhead, not official hardware requirements.
SOURCE URLS
FIELD EVIDENCE
licenseai.google.dev/gemma/terms
architectureai.google.dev/gemma/docs/core/model_card_2
contextWindowTokensdevelopers.googleblog.com/gemma-explained-overview-gemma-model-family-architectures/
trainingTokensai.google.dev/gemma/docs/core/model_card_2
releaseDateai.google.dev/gemma/docs/releases
hfGgufReposhuggingface.co/bartowski/gemma-2-2b-it-GGUF
q4FileSizeGbhuggingface.co/bartowski/gemma-2-2b-it-GGUF
q5FileSizeGbhuggingface.co/bartowski/gemma-2-2b-it-GGUF
q8FileSizeGbhuggingface.co/bartowski/gemma-2-2b-it-GGUF
hardwareEstimateshuggingface.co/bartowski/gemma-2-2b-it-GGUF
LANGUAGES
English
CAPABILITIES
No special capabilities flagged.