52 lines
2.6 KiB
Markdown
52 lines
2.6 KiB
Markdown
|
---
|
||
|
|
license: apache-2.0
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
base_model:
|
||
|
|
- prithivMLmods/Blitzar-Coder-4B-F.1
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
library_name: transformers
|
||
|
|
tags:
|
||
|
|
- text-generation-inference
|
||
|
|
- code
|
||
|
|
- coder
|
||
|
|
---
|
||
|
|
|
||
|
|
# **Blitzar-Coder-4B-F.1-GGUF**
|
||
|
|
|
||
|
|
> **Blitzar-Coder-4B-F.1** is a high-efficiency, multi-language coding model fine-tuned on **Qwen3-4B** using **larger coding traces datasets** spanning **10+ programming languages** including Python, Java, C#, C++, C, Go, JavaScript, TypeScript, Rust, and more. This model delivers exceptional code generation, debugging, and reasoning capabilities—making it an ideal tool for developers seeking advanced programming assistance under constrained compute.
|
||
|
|
|
||
|
|
## Model Files
|
||
|
|
|
||
|
|
| Filename | Size | Format | Description |
|
||
|
|
|----------|------|--------|-------------|
|
||
|
|
| Blitzar-Coder-4B-F.1.BF16.gguf | 8.05 GB | BF16 | Brain Float 16-bit quantization |
|
||
|
|
| Blitzar-Coder-4B-F.1.F16.gguf | 8.05 GB | F16 | Half precision (16-bit) floating point |
|
||
|
|
| Blitzar-Coder-4B-F.1.F32.gguf | 16.1 GB | F32 | Full precision (32-bit) floating point |
|
||
|
|
| Blitzar-Coder-4B-F.1.Q2_K.gguf | 1.67 GB | Q2_K | 2-bit quantization with K-quant |
|
||
|
|
| Blitzar-Coder-4B-F.1.Q3_K_L.gguf | 2.24 GB | Q3_K_L | 3-bit quantization (Large) with K-quant |
|
||
|
|
| Blitzar-Coder-4B-F.1.Q3_K_M.gguf | 2.08 GB | Q3_K_M | 3-bit quantization (Medium) with K-quant |
|
||
|
|
| Blitzar-Coder-4B-F.1.Q3_K_S.gguf | 1.89 GB | Q3_K_S | 3-bit quantization (Small) with K-quant |
|
||
|
|
| Blitzar-Coder-4B-F.1.Q4_K_M.gguf | 2.5 GB | Q4_K_M | 4-bit quantization (Medium) with K-quant |
|
||
|
|
| Blitzar-Coder-4B-F.1.Q4_K_S.gguf | 2.38 GB | Q4_K_S | 4-bit quantization (Small) with K-quant |
|
||
|
|
| Blitzar-Coder-4B-F.1.Q5_K_M.gguf | 2.89 GB | Q5_K_M | 5-bit quantization (Medium) with K-quant |
|
||
|
|
| Blitzar-Coder-4B-F.1.Q5_K_S.gguf | 2.82 GB | Q5_K_S | 5-bit quantization (Small) with K-quant |
|
||
|
|
| Blitzar-Coder-4B-F.1.Q6_K.gguf | 3.31 GB | Q6_K | 6-bit quantization with K-quant |
|
||
|
|
| Blitzar-Coder-4B-F.1.Q8_0.gguf | 4.28 GB | Q8_0 | 8-bit quantization |
|
||
|
|
|
||
|
|
### Recommended Usage
|
||
|
|
|
||
|
|
- **Q4_K_M** or **Q5_K_M**: Best balance of quality and performance for most users
|
||
|
|
- **Q6_K** or **Q8_0**: Higher quality, larger file sizes
|
||
|
|
- **Q2_K** or **Q3_K_S**: Fastest inference, lower quality
|
||
|
|
- **F16** or **BF16**: High quality, requires more VRAM
|
||
|
|
- **F32**: Highest quality, requires significant VRAM
|
||
|
|
|
||
|
|
## 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):
|
||
|
|
|
||
|
|
|