Capella-Qwen3-DS-V3.1-4B-GGUF

Capella-Qwen3-DS-V3.1-4B is a reasoning-focused model fine-tuned on Qwen-4B using DeepSeek v3.1 synthetic traces (10K). It specializes in random event simulations, logical problem analysis, and structured reasoning tasks. The model blends symbolic precision, probabilistic logic, and structured output fluency—making it an ideal tool for researchers, educators, and developers working with uncertainty modeling and event-driven analysis.

Model Files

File Name	Quant Type	File Size
Capella-Qwen3-DS-V3.1-4B.BF16.gguf	BF16	8.05 GB
Capella-Qwen3-DS-V3.1-4B.F16.gguf	F16	8.05 GB
Capella-Qwen3-DS-V3.1-4B.F32.gguf	F32	16.1 GB
Capella-Qwen3-DS-V3.1-4B.Q2_K.gguf	Q2_K	1.67 GB
Capella-Qwen3-DS-V3.1-4B.Q3_K_L.gguf	Q3_K_L	2.24 GB
Capella-Qwen3-DS-V3.1-4B.Q3_K_M.gguf	Q3_K_M	2.08 GB
Capella-Qwen3-DS-V3.1-4B.Q3_K_S.gguf	Q3_K_S	1.89 GB
Capella-Qwen3-DS-V3.1-4B.Q4_K_M.gguf	Q4_K_M	2.5 GB
Capella-Qwen3-DS-V3.1-4B.Q4_K_S.gguf	Q4_K_S	2.38 GB
Capella-Qwen3-DS-V3.1-4B.Q5_K_M.gguf	Q5_K_M	2.89 GB
Capella-Qwen3-DS-V3.1-4B.Q5_K_S.gguf	Q5_K_S	2.82 GB
Capella-Qwen3-DS-V3.1-4B.Q6_K.gguf	Q6_K	3.31 GB
Capella-Qwen3-DS-V3.1-4B.Q8_0.gguf	Q8_0	4.28 GB

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):