Modelpublic

bartowski/c4ai-command-r-plus-GGUF

sourceHugging Facecc-by-nc-4.0updated 2y agoView on Hugging Face

7likes1.7kdownloads

Model Card

Llamacpp Quantizations of c4ai-command-r-plus

Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b2636">b2636</a> for quantization.

Original model: https://huggingface.co/CohereForAI/c4ai-command-r-plus

Prompt format

<BOS_TOKEN><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>{system_prompt}<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>{prompt}<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|><|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>

Download a file (not the whole branch) from below:

Filename	Quant type	File Size	Description
c4ai-command-r-plus-Q5_K_M.gguf	Q5KM	73.62GB	High quality, recommended.
c4ai-command-r-plus-Q5_K_S.gguf	Q5KS	71.80GB	High quality, recommended.
c4ai-command-r-plus-Q4_K_M.gguf	Q4KM	62.75GB	Good quality, uses about 4.83 bits per weight, recommended.
c4ai-command-r-plus-Q4_K_S.gguf	Q4KS	59.64GB	Slightly lower quality with more space savings, recommended.
c4ai-command-r-plus-IQ4_NL.gguf	IQ4_NL	59.73GB	Decent quality, slightly smaller than Q4KS with similar performance recommended.
c4ai-command-r-plus-IQ4_XS.gguf	IQ4_XS	56.72GB	Decent quality, smaller than Q4KS with similar performance, recommended.
c4ai-command-r-plus-Q3_K_L.gguf	Q3KL	55.40GB	Lower quality but usable, good for low RAM availability.
c4ai-command-r-plus-Q3_K_M.gguf	Q3KM	50.98GB	Even lower quality.
c4ai-command-r-plus-IQ3_M.gguf	IQ3_M	47.68GB	Medium-low quality, new method with decent performance comparable to Q3KM.
c4ai-command-r-plus-IQ3_S.gguf	IQ3_S	45.95GB	Lower quality, new method with decent performance, recommended over Q3KS quant, same size with better performance.
c4ai-command-r-plus-Q3_K_S.gguf	Q3KS	45.85GB	Low quality, not recommended.
c4ai-command-r-plus-IQ3_XS.gguf	IQ3_XS	43.59GB	Lower quality, new method with decent performance, slightly better than Q3KS.
c4ai-command-r-plus-Q2_K.gguf	Q2_K	39.49GB	Very low quality but surprisingly usable.

Which file should I choose?

A great write up with charts showing various performances is provided by Artefact2 here

The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.

If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.

If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.

Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.

If you don't want to think too much, grab one of the K-quants. These are in format 'QXKX', like Q5KM.

If you want to get more into the weeds, you can check out this extremely useful feature chart:

llama.cpp feature matrix

But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQXX, like IQ3M. These are newer and offer better performance for their size.

These I-quants can also be used on CPU and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.

The I-quants are not compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski