by mradermacher
Open source · 436 downloads · 4 likes
The "zen musician i1 GGUF" model is an optimized and quantized version of a specialized AI designed for music creation and generating text related to music. It excels in composing melodies, writing lyrics, or generating musical descriptions, while delivering fluidity and creativity. Its use cases include assisting musicians with composition, generating educational music content, or inspiring artistic projects. What sets it apart is its ability to balance technical precision with artistic sensitivity, thanks to optimized quantizations that strike a balance between performance and quality. The available quantized versions allow it to adapt to diverse needs, from lightweight environments to more demanding setups.
weighted/imatrix quants of https://huggingface.co/zenlm/zen-musician
For a convenient overview and download list, visit our model page for this model.
static quants are available at https://huggingface.co/mradermacher/zen-musician-GGUF
If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files.
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|---|---|---|---|
| GGUF | imatrix | 0.1 | imatrix file (for creating your own quants) |
| GGUF | i1-IQ1_S | 1.6 | for the desperate |
| GGUF | i1-IQ1_M | 1.7 | mostly desperate |
| GGUF | i1-IQ2_XXS | 1.9 | |
| GGUF | i1-IQ2_XS | 2.1 | |
| GGUF | i1-IQ2_S | 2.2 | |
| GGUF | i1-IQ2_M | 2.4 | |
| GGUF | i1-Q2_K_S | 2.4 | very low quality |
| GGUF | i1-Q2_K | 2.5 | IQ3_XXS probably better |
| GGUF | i1-IQ3_XXS | 2.6 | lower quality |
| GGUF | i1-IQ3_XS | 2.8 | |
| GGUF | i1-Q3_K_S | 2.9 | IQ3_XS probably better |
| GGUF | i1-IQ3_S | 2.9 | beats Q3_K* |
| GGUF | i1-IQ3_M | 3.0 | |
| GGUF | i1-Q3_K_M | 3.2 | IQ3_S probably better |
| GGUF | i1-Q3_K_L | 3.4 | IQ3_M probably better |
| GGUF | i1-IQ4_XS | 3.5 | |
| GGUF | i1-IQ4_NL | 3.7 | prefer IQ4_XS |
| GGUF | i1-Q4_0 | 3.7 | fast, low quality |
| GGUF | i1-Q4_K_S | 3.7 | optimal size/speed/quality |
| GGUF | i1-Q4_K_M | 3.9 | fast, recommended |
| GGUF | i1-Q4_1 | 4.1 | |
| GGUF | i1-Q5_K_S | 4.4 | |
| GGUF | i1-Q5_K_M | 4.5 | |
| GGUF | i1-Q6_K | 5.2 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):
And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized.
I thank my company, nethype GmbH, for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
