Run ggml-org/embeddinggemma-300M-GGUF locally

⬇ 1,377,020 ❤ 33
Parameters0.31B
Context2,048

ggml-org/embeddinggemma-300M-GGUF is a compact language model with 0.31 billion parameters, built on the gemma-embedding architecture. It has been downloaded 1,377,020 times.

To run ggml-org/embeddinggemma-300M-GGUF locally at a 4,096-token context, its quantized versions need between 1.2 GB (Q8_0, lowest quality) and 1.2 GB (Q8_0, highest quality) of memory, weights plus KV cache and a system margin included.

For most users the best balance is Q8_0, needing about 1.2 GB. That means ggml-org/embeddinggemma-300M-GGUF fits entirely in the VRAM of a 6 GB GPU or larger, running fully on the GPU.

→ Guide: How much VRAM do you need?

All quantizations

Quant.Bits QualityWeights KVTotal Speed~Verdict
Q8_0 8.68 Excellent 0.31 GB 0.09 GB 1.2 GB 1287.5 t/s Fits in VRAM

KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.

Frequently asked questions

How much VRAM do you need to run ggml-org/embeddinggemma-300M-GGUF?

You need about 1.2 GB of VRAM to run ggml-org/embeddinggemma-300M-GGUF entirely on the GPU using the Q8_0 quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.

Can I run ggml-org/embeddinggemma-300M-GGUF on an 8 GB GPU?

Yes. With 8 GB of VRAM you can run ggml-org/embeddinggemma-300M-GGUF fully on the GPU using Q8_0 (about 1.2 GB).

Can I run ggml-org/embeddinggemma-300M-GGUF on a 16 GB GPU?

Yes. With 16 GB of VRAM you can run ggml-org/embeddinggemma-300M-GGUF fully on the GPU using Q8_0 (about 1.2 GB).

Can I run ggml-org/embeddinggemma-300M-GGUF on a 24 GB GPU?

Yes. With 24 GB of VRAM you can run ggml-org/embeddinggemma-300M-GGUF fully on the GPU using Q8_0 (about 1.2 GB).

What is the best quantization for ggml-org/embeddinggemma-300M-GGUF?

If memory allows, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.