GE

malaysian-gemma-2b-16k-instructions

GemmaGGUF by prsyahmiUpdated 2024-04-01
Download on Hugging Face
Parameters2.51B
Context32K
Layers18
Architecturegemma
Smallest quant1.2 GB
Quick answer

malaysian-gemma-2b-16k-instructions needs about 2.3 GB of memory at Q4_K_M (a 1.6 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 8 quants, from 1.2 GB (Q2_K) to 10.0 GB (F32). Longer context needs more: at 32K tokens the KV cache alone adds about 0.6 GB.

malaysian-gemma-2b-16k-instructions is a 2.51B-parameter Gemma model in GGUF format, quantized by prsyahmi. Its trained context length is 32K tokens. The repository has 8 quant files, from 1.2 GB (Q2_K) to 10.0 GB (F32).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K1.2 GB 1.8 GB
Fits Get file →
Q3_K_L1.5 GB 2.1 GB
Fits Get file →
Q4_K_M1.6 GB 2.3 GB
Fits Get file →
Q5_K_M1.8 GB 2.5 GB
Fits Get file →
Q6_K2.1 GB 2.7 GB
Fits Get file →
Q8_02.7 GB 3.3 GB
Fits Get file →
F165.0 GB 5.7 GB
Fits Get file →
F3210.0 GB 10.7 GB
Too big Get file →

What GPU do you need for malaysian-gemma-2b-16k-instructions?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is Q8_0 (2.7 GB file, about 3.3 GB in use).
  • 6 GB VRAM: the best quant that fits is F16 (5.0 GB file, about 5.7 GB in use).
  • 8 GB VRAM: the best quant that fits is F16 (5.0 GB file, about 5.7 GB in use).
  • 12 GB VRAM: the best quant that fits is F32 (10.0 GB file, about 10.7 GB in use).

Which malaysian-gemma-2b-16k-instructions quant should you choose?

Q4_K_M is the usual starting point for malaysian-gemma-2b-16k-instructions: it balances file size and quality well. Here it is a 1.6 GB file.

The Q8_0 file is 2.7 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q2_K at 1.2 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturegemma
Parameters2.51B
Layers18
Trained context32,768 tokens
KV cache (fp16)18 KB per token

Source

Quantized byprsyahmi
FamilyGemma
Downloads (HF)429
Last updated2024-04-01

Frequently asked questions about malaysian-gemma-2b-16k-instructions

How much VRAM does malaysian-gemma-2b-16k-instructions need?

About 2.3 GB at Q4_K_M with an 8K context window. That is the 1.6 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run malaysian-gemma-2b-16k-instructions on an 8 GB GPU?

Yes, with F16: the 5.0 GB file needs about 5.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of malaysian-gemma-2b-16k-instructions GGUF?

It depends on the quant: Q2_K is 1.2 GB, Q3_K_L is 1.5 GB, Q4_K_M is 1.6 GB, Q5_K_M is 1.8 GB, Q6_K is 2.1 GB, Q8_0 is 2.7 GB and 2 more.

What context length does malaysian-gemma-2b-16k-instructions support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.6 GB at 32K.

Where can I download malaysian-gemma-2b-16k-instructions GGUF?

The files are on Hugging Face in the prsyahmi/malaysian-gemma-2B-16k-instructions-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Similar models

Browse all →

More from prsyahmi

All by prsyahmi →

More Gemma models

All Gemma →

Other tiny models

All tiny models →