mradermacher/Meta-Llama-3.1-405B-Instruct-i1-GGUF

About

weighted/imatrix quants of https://huggingface.co/meta-llama/Meta-Llama-3.1-405B-Instruct

These imatrix quants have recently been requantized using a higher quality imatrix calculated directly from the source model (instead of the Q8_0 quant used for the previous imatrix). This was probably the largest distributed imatrix computation to date (and also one of the first).

static quants are available at https://huggingface.co/mradermacher/Meta-Llama-3.1-405B-Instruct-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
PART 1 PART 2	i1-IQ1_S	85.3	for the desperate
PART 1 PART 2	i1-IQ1_M	93.6	mostly desperate
PART 1 PART 2 PART 3	i1-IQ2_XXS	107.4
PART 1 PART 2 PART 3	i1-IQ2_XS	119.4
PART 1 PART 2 PART 3	i1-IQ2_S	125.7
PART 1 PART 2 PART 3	i1-IQ2_M	136.8
PART 1 PART 2 PART 3 PART 4	i1-Q2_K	149.4	IQ3_XXS probably better
PART 1 PART 2 PART 3 PART 4	i1-IQ3_XXS	155.9	lower quality
PART 1 PART 2 PART 3 PART 4	i1-IQ3_XS	166.2
PART 1 PART 2 PART 3 PART 4	i1-Q3_K_S	175.3	IQ3_XS probably better
PART 1 PART 2 PART 3 PART 4	i1-IQ3_S	175.6	beats Q3_K*
PART 1 PART 2 PART 3 PART 4	i1-IQ3_M	181.8
PART 1 PART 2 PART 3 PART 4	i1-Q3_K_M	195.5	IQ3_S probably better
P1 P2 P3 P4 P5	i1-Q3_K_L	212.9	IQ3_M probably better
P1 P2 P3 P4 P5	i1-IQ4_XS	216.7
P1 P2 P3 P4 P5	i1-Q4_0	229.8	fast, low quality
P1 P2 P3 P4 P5	i1-Q4_K_S	230.6	optimal size/speed/quality
P1 P2 P3 P4 P5	i1-Q4_K_M	243.2	fast, recommended
P1 P2 P3 P4 P5 P6	i1-Q5_K_S	279.4
P1 P2 P3 P4 P5 P6	i1-Q5_K_M	286.7
P1 P2 P3 P4 P5 P6 P7	i1-Q6_K	333.0	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

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.