DavidAU/L3-Dark-Planet-8B-GGUF

Text Generation • Updated 29 days ago • 1.14k • 38

Note The original Dark Planet in GGUF, with links to new 128k / 1 million context versions as well as expanded Dark Planet versions like DARKEST PLANET 16.5B and other versions too.

DavidAU/L3-Dark-Planet-8B

Text Generation • Updated 24 days ago • 55 • 6

Note Original Dark Planet in full precision / source. Mergekit file included.

DavidAU/L3-MOE-8X8B-Dark-Planet-8D-Mirrored-Chaos-47B-GGUF

Text Generation • Updated 15 days ago • 1.01k • 15

Note The model shows the use of EIGHT of the "Dark Planet Wordstorm" ("cr2", "cr1", "r7", "r6", "b3", "b4", "r1" and "b6") models in a MOE (Mixture of experts) configuration. This allows you to use the power of up to 8 of these models at the same time.

DavidAU/L3.1-MOE-8X8B-Dark-Planet-8D-Mirrored-Chaos-Uncensored-47B-GGUF

Text Generation • Updated 29 days ago • 399 • 2

Note The model shows the use of EIGHT of the "Dark Planet Wordstorm" ("cr2", "cr1", "r7", "r6", "b3", "b4", "r1" and "b6") models in a MOE (Mixture of experts) configuration. This allows you to use the power of up to 8 of these models at the same time. This uses a Llama 3.1 model as a "base" in the MOE to extend context to 128k.

DavidAU/L3-MOE-4x8B-Dark-Planet-Rising-25B-GGUF

Text Generation • Updated 29 days ago • 574 • 4

Note The model shows the use of FOUR of the "Dark Planet Wordstorm" ("cr2","cr1","r7" and "r6") models in a MOE (Mixture of experts) configuration. This allows you to use the power of up to 4 of these models at the same time.

DavidAU/L3-MOE-4x8B-Dark-Planet-Rebel-FURY-25B-GGUF

Text Generation • Updated 29 days ago • 530 • 3

Note The model shows the use of FOUR of the "Dark Planet Wordstorm" ("b3","b4","r1" and "b6") models in a MOE (Mixture of experts) configuration. This allows you to use the power of up to 4 of these models at the same time.

DavidAU/Llama-3.1-Dark-Planet-8B-SuperNova

Text Generation • Updated 23 days ago • 98

Note Use of the exact Dark Planet 8B original formula, except the "base" model (Meta-Llama Instruct Llama 3) was replaced with Llama 3.1 Super Nova 8B. This alters both generation output, and changes the context maximum from 8k to 128k as Super Nova is a LLama 3.1 model. Mergekit file is listed in the "files" section.

DavidAU/Llama-3.1-Dark-Planet-SuperNova-8B-GGUF

Text Generation • Updated 21 days ago • 2.29k

Note GGUF Quants of LLama 3.1-Dark-Planet-8B-SuperNova

DavidAU/L3-Dark-Planet-8B-wordstorm1

Text Generation • Updated 23 days ago • 107 • 1

Note Base changes, and "density" are used to change the model from Dark Planet original. Wordstorm 1 and 2 are TWO versions showing applied "pruning" caused by using "density". Because of "pruning" wordstorm 1 and 2 produce very different output.

DavidAU/L3-Dark-Planet-8B-wordstorm2

Text Generation • Updated 23 days ago • 90 • 1

Note Base changes, and "density" are used to change the model from Dark Planet original. Wordstorm 1 and 2 are TWO versions showing applied "pruning" caused by using "density". Because of "pruning" wordstorm 1 and 2 produce very different output.

DavidAU/L3-Dark-Planet-8B-wordstorm-b3

Text Generation • Updated 23 days ago • 225 • 1

Note B3, B4 and B6 use the same formula as original Dark Planet 8b, but apply random pruning to each model in merge (not including "base"). These 3 "B series" are the "winners" - best of - for this type of pruning / model merge.

DavidAU/L3-Dark-Planet-8B-wordstorm-b4

Text Generation • Updated 23 days ago • 185 • 1

Note B3, B4 and B6 use the same formula as original Dark Planet 8b, but apply random pruning to each model in merge (not including "base"). These 3 "B series" are the "winners" - best of - for this type of pruning / model merge.

DavidAU/L3-Dark-Planet-8B-wordstorm-b6

Text Generation • Updated 23 days ago • 109 • 1

Note B3, B4 and B6 use the same formula as original Dark Planet 8b, but apply random pruning to each model in merge (not including "base"). These 3 "B series" are the "winners" - best of - for this type of pruning / model merge.

DavidAU/L3-Dark-Planet-8B-wordstorm-cr1

Text Generation • Updated 23 days ago • 54 • 1

Note These merges - CR1/CR2 - uses the original Dark Planet formula, but uses density above 1 for two models and only ONE model with "random pruning" - density at .95 ;

DavidAU/L3-Dark-Planet-8B-wordstorm-cr2

Text Generation • Updated 23 days ago • 108 • 1

Note These merges - CR1/CR2 - uses the original Dark Planet formula, but uses density above 1 for two models and only ONE model with "random pruning" - density at .95 ;

DavidAU/L3-Dark-Planet-8B-wordstorm-r1

Text Generation • Updated 23 days ago • 120 • 1

Note R1, R4, R6, R7 use the same formula as original Dark Planet 8b, but apply random pruning to each model in merge (not including "base"). These 4 "R series" are the "winners" - best of - for this type of pruning / model merge.

DavidAU/L3-Dark-Planet-8B-wordstorm-r4

Text Generation • Updated 23 days ago • 152 • 1

Note R1, R4, R6, R7 use the same formula as original Dark Planet 8b, but apply random pruning to each model in merge (not including "base"). These 4 "R series" are the "winners" - best of - for this type of pruning / model merge.

DavidAU/L3-Dark-Planet-8B-wordstorm-r7

Text Generation • Updated 23 days ago • 198 • 1

Note R1, R4, R6, R7 use the same formula as original Dark Planet 8b, but apply random pruning to each model in merge (not including "base"). These 4 "R series" are the "winners" - best of - for this type of pruning / model merge.

DavidAU/L3-Dark-Planet-8B-wordstorm-r6

Text Generation • Updated 23 days ago • 114 • 1

Note R1, R4, R6, R7 use the same formula as original Dark Planet 8b, but apply random pruning to each model in merge (not including "base"). These 4 "R series" are the "winners" - best of - for this type of pruning / model merge.