mradermacher/Llama-2-70b-x8-MoE-clown-truck-i1-GGUF

About

weighted/imatrix quants of https://huggingface.co/NobodyExistsOnTheInternet/Llama-2-70b-x8-MoE-clown-truck

No IQ1_S is available because llama.cpp is a crashfest atm. and crashes when trying to generate it.

static quants are available at https://huggingface.co/mradermacher/Llama-2-70b-x8-MoE-clown-truck-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 PART 3 PART 4	i1-IQ2_M	153.4
PART 1 PART 2 PART 3 PART 4	i1-Q2_K	170.8	IQ3_XXS probably better
PART 1 PART 2 PART 3 PART 4	i1-IQ3_XXS	180.2	lower quality
PART 1 PART 2 PART 3 PART 4	i1-IQ3_XS	190.5
P1 P2 P3 P4 P5	i1-IQ3_S	202.0	beats Q3_K*
P1 P2 P3 P4 P5	i1-Q3_K_S	202.0	IQ3_XS probably better
P1 P2 P3 P4 P5	i1-IQ3_M	211.9
P1 P2 P3 P4 P5	i1-Q3_K_M	223.1	IQ3_S probably better
P1 P2 P3 P4 P5	i1-Q3_K_L	239.3	IQ3_M probably better
P1 P2 P3 P4 P5 P6	i1-IQ4_XS	248.3
P1 P2 P3 P4 P5 P6	i1-Q4_K_S	264.9	optimal size/speed/quality
P1 P2 P3 P4 P5 P6	i1-Q4_K_M	282.0	fast, recommended
P1 P2 P3 P4 P5 P6 P7	i1-Q5_K_S	319.7
P1 P2 P3 P4 P5 P6 P7	i1-Q5_K_M	329.7
P1 P2 P3 P4 P5 P6 P7 P8	i1-Q6_K	381.0	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.