Written by Jakub Rusinowski · Last updated June 27, 2024
Built from the same research and technology as Google Gemini. Gemma 2 uses knowledge distillation for extremely high efficiency and creative performance.
| Licence | What it permits | Applies to |
|---|---|---|
Gemma Terms | Commercial use permitted Weights are downloadable and commercial use is permitted, subject to the licence’s acceptable-use terms. | Gemma 2 9B IT |
| Gemma 2 9B IT | Min 6 GB VRAM · Q4_K_M · 8,192 ctx · ollama run gemma2 |
The cheapest GPU that runs Gemma 2 Family locally (min 6 GB VRAM) is the Intel Arc B570 (10 GB).
Install Ollama then run: ollama run gemma2
Minimum VRAM: 6 GB. For best results use Q4_K_M quantization.
Pick a quantization and open it in LM Studio, Ollama, or Jan, or download the raw .gguf file directly. Quant list and sizes resolved from Hugging Face.
| Quant | Size | Download (.gguf) |
|---|---|---|
| Q3_K_M | 3.84 GB (est.) | gemma-2-9b-it-Q3_K_M.gguf |
| Q4_K_M | 5.43 GB (est.) | gemma-2-9b-it-Q4_K_M.gguf |
| Q5_K_M | 6.38 GB (est.) | gemma-2-9b-it-Q5_K_M.gguf |
| Q6_K | 7.38 GB (est.) | gemma-2-9b-it-Q6_K.gguf |
| Q8_0 | 9.56 GB (est.) | gemma-2-9b-it-Q8_0.gguf |
Download in LM Studio: lms get bartowski/gemma-2-9b-it-GGUF
Want this model on your phone? You can run it on your desktop with LM Studio and chat from your iPhone or iPad over an encrypted link — see Run LM Studio Models on Your Phone (LM Link).
Gemma 2 Family needs about 6 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Gemma 2 9B IT (6 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Yes — Gemma 2 Family runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
Q4_K_M is the best balance of quality and VRAM for Gemma 2 Family in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
Install Ollama, then run: ollama run gemma2. This downloads Gemma 2 Family and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.