Ross-640-BMath-1.5B-GGUF/README.md

---
license: apache-2.0
language:
- en
base_model:
- prithivMLmods/Ross-640-BMath-1.5B
pipeline_tag: text-generation
library_name: transformers
tags:
- text-generation-inference
- math
---
# **Ross-640-BMath-1.5B-GGUF**

> **Ross-640-BMath-1.5B** is an **experimental, high-precision math explanation model** fine-tuned on **Qwen2-1.5B**, designed to provide **step-by-step mathematical derivations** and **detailed concept explanations** across a wide range of mathematical domains. It is **not optimized for general reasoning or conversation**, and focuses primarily on **structured, non-reasoning math workflows** including algebra, calculus, number theory, and combinatorics.

## Model Files

| File Name | Size | Format | Description |
|-----------|------|--------|-------------|
| Ross-640-BMath-1.5B.F32.gguf | 6.18 GB | F32 | Full precision 32-bit floating point |
| Ross-640-BMath-1.5B.F16.gguf | 3.09 GB | F16 | Half precision 16-bit floating point |
| Ross-640-BMath-1.5B.BF16.gguf | 3.09 GB | BF16 | Brain floating point 16-bit |
| Ross-640-BMath-1.5B.Q8_0.gguf | 1.65 GB | Q8_0 | 8-bit quantized |
| Ross-640-BMath-1.5B.Q6_K.gguf | 1.27 GB | Q6_K | 6-bit quantized |
| Ross-640-BMath-1.5B.Q5_K_M.gguf | 1.13 GB | Q5_K_M | 5-bit quantized, medium quality |
| Ross-640-BMath-1.5B.Q5_K_S.gguf | 1.1 GB | Q5_K_S | 5-bit quantized, small quality |
| Ross-640-BMath-1.5B.Q4_K_M.gguf | 986 MB | Q4_K_M | 4-bit quantized, medium quality |
| Ross-640-BMath-1.5B.Q4_K_S.gguf | 940 MB | Q4_K_S | 4-bit quantized, small quality |
| Ross-640-BMath-1.5B.Q3_K_L.gguf | 880 MB | Q3_K_L | 3-bit quantized, large quality |
| Ross-640-BMath-1.5B.Q3_K_M.gguf | 824 MB | Q3_K_M | 3-bit quantized, medium quality |
| Ross-640-BMath-1.5B.Q3_K_S.gguf | 761 MB | Q3_K_S | 3-bit quantized, small quality |
| Ross-640-BMath-1.5B.Q2_K.gguf | 676 MB | Q2_K | 2-bit quantized |

## Quants Usage 

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

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

![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)