Basically-Human-4B-f32-GGUF/README.md

license, language, base_model, library_name, pipeline_tag, tags

license	language	base_model	library_name	pipeline_tag	tags
apache-2.0	en	GreenerPastures/Basically-Human-4B	transformers	text-generation	text-generation-inference

Basically-Human-4B-f32-GGUF

Basically-Human-4B is a 4 billion parameter language model based on the Qwen3 4B architecture, fine-tuned specifically for immersive and emotionally resonant roleplaying and character interaction. It excels in maintaining in-character consistency, crafting believable dialogue, and driving dynamic storytelling, making it ideal for text-based roleplay, NPC simulation, and interactive fiction applications. The model uses the ChatML instruction format to structure multi-turn conversations with clear role delineation. Basically-Human-4B was fine-tuned on a diverse set of instruction and roleplaying datasets, including various curated and cleaned instruction data from multiple sources. It offers GGUF quantized versions for ease of deployment while delivering compact yet capable performance in roleplay scenarios.

Model Files

File Name	Quant Type	File Size
Basically-Human-4B.BF16.gguf	BF16	8.05 GB
Basically-Human-4B.F16.gguf	F16	8.05 GB
Basically-Human-4B.F32.gguf	F32	16.1 GB
Basically-Human-4B.Q2_K.gguf	Q2_K	1.67 GB
Basically-Human-4B.Q3_K_L.gguf	Q3_K_L	2.24 GB
Basically-Human-4B.Q3_K_M.gguf	Q3_K_M	2.08 GB
Basically-Human-4B.Q3_K_S.gguf	Q3_K_S	1.89 GB
Basically-Human-4B.Q4_K_M.gguf	Q4_K_M	2.5 GB
Basically-Human-4B.Q4_K_S.gguf	Q4_K_S	2.38 GB
Basically-Human-4B.Q5_K_M.gguf	Q5_K_M	2.89 GB
Basically-Human-4B.Q5_K_S.gguf	Q5_K_S	2.82 GB
Basically-Human-4B.Q6_K.gguf	Q6_K	3.31 GB
Basically-Human-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):