mradermacher/Triangulum-10B-i1-GGUF

---

base_model: prithivMLmods/Triangulum-10B language:

• en
• de
• fr
• it
• pt
• hi
• es
• th

library_name: transformers license: llama3.1 quantized_by: mradermacher tags:

• triangulum10b
• sft
• chainofthought
• ollama
• text-generation-inference
• llamaforcausallm
• reasoning
• CoT

---

About

weighted/imatrix quants of https://huggingface.co/prithivMLmods/Triangulum-10B

static quants are available at https://huggingface.co/mradermacher/Triangulum-10B-GGUF

Usage

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.

Provided Quants

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

Link	Type	Size/GB	Notes
GGUF	i1-IQ1_S	2.5	for the desperate
GGUF	i1-IQ1_M	2.7	mostly desperate
GGUF	i1-IQ2_XXS	3.0
GGUF	i1-IQ2_XS	3.3
GGUF	i1-IQ2_S	3.4
GGUF	i1-IQ2_M	3.7
GGUF	i1-Q2K_S	3.8	very low quality
GGUF	i1-Q2K	4.0	IQ3XXS probably better
GGUF	i1-IQ3_XXS	4.2	lower quality
GGUF	i1-IQ3_XS	4.5
GGUF	i1-IQ3S	4.7	beats Q3K*
GGUF	i1-Q3KS	4.7	IQ3XS probably better
GGUF	i1-IQ3_M	4.8
GGUF	i1-Q3KM	5.2	IQ3S probably better
GGUF	i1-Q3KL	5.6	IQ3M probably better
GGUF	i1-IQ4_XS	5.7
GGUF	i1-IQ4NL	6.0	prefer IQ4XS
GGUF	i1-Q4_0	6.0	fast, low quality
GGUF	i1-Q4K_S	6.1	optimal size/speed/quality
GGUF	i1-Q4K_M	6.4	fast, recommended
GGUF	i1-Q4_1	6.6
GGUF	i1-Q5K_S	7.2
GGUF	i1-Q5K_M	7.4
GGUF	i1-Q6K	8.6	practically like static Q6K

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):

!image.png

And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9

FAQ / Model Request

See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.

Thanks

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.