CoolFace
Modelpublic

bartowski/Chat2DB-SQL-7B-GGUF

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face
3likes207downloads
Model Card

Llamacpp Quantizations of Chat2DB-SQL-7B

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/Chat2DB/Chat2DB-SQL-7B

All quants made using imatrix option with dataset provided by Kalomaze here

Prompt format

No chat template specified so default is used. This may be incorrect, check original model card for details.

<s> [INST] <<SYS>>
{system_prompt}
<</SYS>>

{prompt} [/INST]  </s>

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

FilenameQuant typeFile SizeDescription
Chat2DB-SQL-7B-Q8_0.ggufQ8_07.16GBExtremely high quality, generally unneeded but max available quant.
Chat2DB-SQL-7B-Q6_K.ggufQ6_K5.52GBVery high quality, near perfect, recommended.
Chat2DB-SQL-7B-Q5_K_M.ggufQ5KM4.78GBHigh quality, recommended.
Chat2DB-SQL-7B-Q5_K_S.ggufQ5KS4.65GBHigh quality, recommended.
Chat2DB-SQL-7B-Q4_K_M.ggufQ4KM4.08GBGood quality, uses about 4.83 bits per weight, recommended.
Chat2DB-SQL-7B-Q4_K_S.ggufQ4KS3.85GBSlightly lower quality with more space savings, recommended.
Chat2DB-SQL-7B-IQ4_NL.ggufIQ4_NL3.82GBDecent quality, slightly smaller than Q4KS with similar performance recommended.
Chat2DB-SQL-7B-IQ4_XS.ggufIQ4_XS3.61GBDecent quality, smaller than Q4KS with similar performance, recommended.
Chat2DB-SQL-7B-Q3_K_L.ggufQ3KL3.59GBLower quality but usable, good for low RAM availability.
Chat2DB-SQL-7B-Q3_K_M.ggufQ3KM3.29GBEven lower quality.
Chat2DB-SQL-7B-IQ3_M.ggufIQ3_M3.11GBMedium-low quality, new method with decent performance comparable to Q3KM.
Chat2DB-SQL-7B-IQ3_S.ggufIQ3_S2.94GBLower quality, new method with decent performance, recommended over Q3KS quant, same size with better performance.
Chat2DB-SQL-7B-Q3_K_S.ggufQ3KS2.94GBLow quality, not recommended.
Chat2DB-SQL-7B-IQ3_XS.ggufIQ3_XS2.79GBLower quality, new method with decent performance, slightly better than Q3KS.
Chat2DB-SQL-7B-IQ3_XXS.ggufIQ3_XXS2.58GBLower quality, new method with decent performance, comparable to Q3 quants.
Chat2DB-SQL-7B-Q2_K.ggufQ2_K2.53GBVery low quality but surprisingly usable.
Chat2DB-SQL-7B-IQ2_M.ggufIQ2_M2.35GBVery low quality, uses SOTA techniques to also be surprisingly usable.
Chat2DB-SQL-7B-IQ2_S.ggufIQ2_S2.19GBVery low quality, uses SOTA techniques to be usable.
Chat2DB-SQL-7B-IQ2_XS.ggufIQ2_XS2.03GBVery low quality, uses SOTA techniques to be usable.
Chat2DB-SQL-7B-IQ2_XXS.ggufIQ2_XXS1.85GBLower quality, uses SOTA techniques to be usable.
Chat2DB-SQL-7B-IQ1_M.ggufIQ1_M1.65GBExtremely low quality, not recommended.
Chat2DB-SQL-7B-IQ1_S.ggufIQ1_S1.52GBExtremely low quality, not recommended.

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