CoolFace
Modelpublic

prithivMLmods/Open-Xi-Math-Preview-GGUF

sourceHugging Faceapache-2.0updated 1y agoView on Hugging Face
0likes345downloads
Model Card

Open-Xi-Math-Preview-GGUF

Open-Xi-Math-Preview is a mathematics-focused reasoning model fine-tuned on Qwen2-1.5B-Instruct, utilizing a modular dataset designed for enhancing mathematical thinking. It provides robust capabilities in symbolic reasoning, structured deduction, and compact coding — optimized for edge deployment on resource-constrained devices.

Model Files

File NameSizeQuantizationFormatDescription
Open-Xi-Math-Preview.BF16.gguf3.56 GBBF16GGUFBFloat16 precision version
Open-Xi-Math-Preview.F16.gguf3.56 GBFP16GGUFFloat16 precision version
Open-Xi-Math-Preview.F32.gguf7.11 GBFP32GGUFFloat32 precision version
Open-Xi-Math-Preview.Q2_K.gguf753 MBQ2\_KGGUF2-bit quantized (K variant)
Open-Xi-Math-Preview.Q3_K_M.gguf924 MBQ3\K\MGGUF3-bit quantized (K M variant)
Open-Xi-Math-Preview.Q4_K_M.gguf1.12 GBQ4\K\MGGUF4-bit quantized (K M variant)
Open-Xi-Math-Preview.Q4_K_S.gguf1.07 GBQ4\K\SGGUF4-bit quantized (K S variant)
Open-Xi-Math-Preview.Q5_K_M.gguf1.29 GBQ5\K\MGGUF5-bit quantized (K M variant)
Open-Xi-Math-Preview.Q5_K_S.gguf1.26 GBQ5\K\SGGUF5-bit quantized (K S variant)
Open-Xi-Math-Preview.Q6_K.gguf1.46 GBQ6\_KGGUF6-bit quantized (K variant)
Open-Xi-Math-Preview.Q8_0.gguf1.89 GBQ8\_0GGUF8-bit quantized
.gitattributes2.39 kB——Git LFS tracking file
config.json31 B——Configuration file
README.md4.29 kB——Model documentation

Quants Usage

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

LinkTypeSize/GBNotes
GGUFQ2_K0.4
GGUFQ3KS0.5
GGUFQ3KM0.5lower quality
GGUFQ3KL0.5
GGUFIQ4_XS0.6
GGUFQ4KS0.6fast, recommended
GGUFQ4KM0.6fast, recommended
GGUFQ5KS0.6
GGUFQ5KM0.7
GGUFQ6_K0.7very good quality
GGUFQ8_00.9fast, best quality
GGUFf161.616 bpw, overkill

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):

image.png