prithivMLmods/Open-Xi-Math-Preview-GGUF

1---2license: apache-2.03language:4- en5base_model:6- prithivMLmods/Open-Xi-Math-Preview7pipeline_tag: text-generation8library_name: transformers9tags:10- text-generation-inference11- math12---13 14# **Open-Xi-Math-Preview-GGUF**15 16> **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**.17 18## Model Files19 20| File Name                          | Size    | Quantization | Format | Description                   |21| ---------------------------------- | ------- | ------------ | ------ | ----------------------------- |22| `Open-Xi-Math-Preview.BF16.gguf`   | 3.56 GB | BF16         | GGUF   | BFloat16 precision version    |23| `Open-Xi-Math-Preview.F16.gguf`    | 3.56 GB | FP16         | GGUF   | Float16 precision version     |24| `Open-Xi-Math-Preview.F32.gguf`    | 7.11 GB | FP32         | GGUF   | Float32 precision version     |25| `Open-Xi-Math-Preview.Q2_K.gguf`   | 753 MB  | Q2\_K        | GGUF   | 2-bit quantized (K variant)   |26| `Open-Xi-Math-Preview.Q3_K_M.gguf` | 924 MB  | Q3\_K\_M     | GGUF   | 3-bit quantized (K M variant) |27| `Open-Xi-Math-Preview.Q4_K_M.gguf` | 1.12 GB | Q4\_K\_M     | GGUF   | 4-bit quantized (K M variant) |28| `Open-Xi-Math-Preview.Q4_K_S.gguf` | 1.07 GB | Q4\_K\_S     | GGUF   | 4-bit quantized (K S variant) |29| `Open-Xi-Math-Preview.Q5_K_M.gguf` | 1.29 GB | Q5\_K\_M     | GGUF   | 5-bit quantized (K M variant) |30| `Open-Xi-Math-Preview.Q5_K_S.gguf` | 1.26 GB | Q5\_K\_S     | GGUF   | 5-bit quantized (K S variant) |31| `Open-Xi-Math-Preview.Q6_K.gguf`   | 1.46 GB | Q6\_K        | GGUF   | 6-bit quantized (K variant)   |32| `Open-Xi-Math-Preview.Q8_0.gguf`   | 1.89 GB | Q8\_0        | GGUF   | 8-bit quantized               |33| `.gitattributes`                   | 2.39 kB | —            | —      | Git LFS tracking file         |34| `config.json`                      | 31 B    | —            | —      | Configuration file            |35| `README.md`                        | 4.29 kB | —            | —      | Model documentation           |36 37## Quants Usage 38 39(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)40 41| Link | Type | Size/GB | Notes |42|:-----|:-----|--------:|:------|43| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.Q2_K.gguf) | Q2_K | 0.4 |  |44| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.Q3_K_S.gguf) | Q3_K_S | 0.5 |  |45| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.Q3_K_M.gguf) | Q3_K_M | 0.5 | lower quality |46| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.Q3_K_L.gguf) | Q3_K_L | 0.5 |  |47| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.IQ4_XS.gguf) | IQ4_XS | 0.6 |  |48| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.Q4_K_S.gguf) | Q4_K_S | 0.6 | fast, recommended |49| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.Q4_K_M.gguf) | Q4_K_M | 0.6 | fast, recommended |50| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.Q5_K_S.gguf) | Q5_K_S | 0.6 |  |51| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.Q5_K_M.gguf) | Q5_K_M | 0.7 |  |52| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.Q6_K.gguf) | Q6_K | 0.7 | very good quality |53| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.Q8_0.gguf) | Q8_0 | 0.9 | fast, best quality |54| [GGUF](https://huggingface.co/mradermacher/Qwen3-0.6B-GGUF/resolve/main/Qwen3-0.6B.f16.gguf) | f16 | 1.6 | 16 bpw, overkill |55 56Here is a handy graph by ikawrakow comparing some lower-quality quant57types (lower is better):58 59![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)