Gemma 2

gemma-2-9b-it

Google DeepMind · open-restricted gemma

Parameters (B)9.242
Architecturedense
Layers / KV heads / head dim42 / 8 / 256
Context window8192
Modalitiesin: text; out: text
GGUFavailable
Commercial useyes

Minimum hardware (estimate)

At Q4_K_M and 8k context, total working set ≈ 9.0 GiB. On a machine with 64 GiB system RAM and no discrete GPU: CPU-only, slow. Limiting factor: No GPU VRAM; fits in system RAM but inference will be CPU-bound.

Open in fit advisor

Scores

BenchmarkValueUncertaintyRun bySource
GPQA Diamond27.5%unknownthird-partyepoch-ai
MATH Level 521.0%unknownthird-partyepoch-ai
OTIS Mock AIME 2024-20250.6%unknownthird-partyepoch-ai

Provenance

HF: google/gemma-2-9b-it · snapshot 2026-08-10