GGUFs that run on one NVIDIA DGX Spark / GB10 (128 GB unified memory) with Abyss: GLM-5.3-Flash, DeepSeek V4 0731, Qwen3.8 (BETA).
zephel01
zephel01
AI & ML interests
None yet
Recent Activity
updated a collection 1 day ago
Ornith-1.5 GGUF updated a model 1 day ago
zephel01/Ornith-1.5-397B-IQ2_XXS-16G64G-GGUF published a model 1 day ago
zephel01/Ornith-1.5-397B-IQ2_XXS-16G64G-GGUFOrganizations
None yet
DeepSeek-V4-Flash-0731 GGUFs (BETA). 32G96G = 32 GB GPU + 96 GB RAM. 32G96G, 32G64G, 16G64G, DGX Spark. Runs on Abyss.
-
zephel01/DeepSeek-V4-Flash-0731-IQ2_M-32G96G-GGUF
Text Generation • 284B • Updated • 40 -
zephel01/DeepSeek-V4-Flash-0731-DGX-Spark-GGUF
Text Generation • 284B • Updated • 34 -
zephel01/DeepSeek-V4-Flash-0731-IQ2_XXS-32G64G-GGUF
Text Generation • 284B • Updated • 36 -
zephel01/DeepSeek-V4-Flash-0731-IQ1_S-16G64G-GGUF
Text Generation • 284B • Updated • 41
DeepSeek-V4.1-Flash-1017 GGUFs (BETA). 32G96G = 32 GB GPU + 96 GB RAM; E256 = 256 of 384 experts. Japanese checked; runs on Abyss.
Ornith-1.5 GGUFs (BETA). 16G64G = 16 GB GPU + 64 GB RAM. 122B: 16G64G, 24G64G; 397B for 32G128G / DGX Spark planned. Runs on Abyss.
-
zephel01/Ornith-1.5-122B-IQ3_XXS-16G64G-GGUF
Text Generation • 125B • Updated • 5.82k • 1 -
zephel01/Ornith-1.5-122B-IQ4_XS-24G64G-GGUF
Text Generation • 125B • Updated • 86 -
zephel01/Ornith-1.5-397B-IQ2_XXS-32G128G-GGUF
Text Generation • 403B • Updated • 5.39k -
zephel01/Ornith-1.5-397B-IQ2_XXS-16G64G-GGUF
Text Generation • 403B • Updated • 38
GLM-5.3-Flash GGUFs (BETA). 32G96G = 32 GB GPU + 96 GB RAM. 32G96G, 32G128G, 24G64G (soon), DGX Spark. Runs on Abyss.
Qwen3.8-Flash-Next GGUFs (BETA). 16G64G = 16 GB GPU + 64 GB RAM; E448 = 448 of 512 experts. Japanese checked; Abyss and llama.cpp.
-
zephel01/Qwen3.8-Flash-Next-E448-IQ2_M-12G32G-GGUF
Text Generation • 162B • Updated • 223 • 5 -
zephel01/Qwen3.8-Flash-Next-IQ3_M-16G64G-GGUF
Text Generation • 177B • Updated • 129 • 1 -
zephel01/Qwen3.8-Flash-Next-IQ3_M-24G64G-GGUF
Text Generation • 177B • Updated • 85 • 1 -
zephel01/Qwen3.8-Flash-Next-DGX-Spark-GGUF
Text Generation • 177B • Updated • 35
GGUFs that run on one NVIDIA DGX Spark / GB10 (128 GB unified memory) with Abyss: GLM-5.3-Flash, DeepSeek V4 0731, Qwen3.8 (BETA).
Ornith-1.5 GGUFs (BETA). 16G64G = 16 GB GPU + 64 GB RAM. 122B: 16G64G, 24G64G; 397B for 32G128G / DGX Spark planned. Runs on Abyss.
-
zephel01/Ornith-1.5-122B-IQ3_XXS-16G64G-GGUF
Text Generation • 125B • Updated • 5.82k • 1 -
zephel01/Ornith-1.5-122B-IQ4_XS-24G64G-GGUF
Text Generation • 125B • Updated • 86 -
zephel01/Ornith-1.5-397B-IQ2_XXS-32G128G-GGUF
Text Generation • 403B • Updated • 5.39k -
zephel01/Ornith-1.5-397B-IQ2_XXS-16G64G-GGUF
Text Generation • 403B • Updated • 38
DeepSeek-V4-Flash-0731 GGUFs (BETA). 32G96G = 32 GB GPU + 96 GB RAM. 32G96G, 32G64G, 16G64G, DGX Spark. Runs on Abyss.
-
zephel01/DeepSeek-V4-Flash-0731-IQ2_M-32G96G-GGUF
Text Generation • 284B • Updated • 40 -
zephel01/DeepSeek-V4-Flash-0731-DGX-Spark-GGUF
Text Generation • 284B • Updated • 34 -
zephel01/DeepSeek-V4-Flash-0731-IQ2_XXS-32G64G-GGUF
Text Generation • 284B • Updated • 36 -
zephel01/DeepSeek-V4-Flash-0731-IQ1_S-16G64G-GGUF
Text Generation • 284B • Updated • 41
GLM-5.3-Flash GGUFs (BETA). 32G96G = 32 GB GPU + 96 GB RAM. 32G96G, 32G128G, 24G64G (soon), DGX Spark. Runs on Abyss.
DeepSeek-V4.1-Flash-1017 GGUFs (BETA). 32G96G = 32 GB GPU + 96 GB RAM; E256 = 256 of 384 experts. Japanese checked; runs on Abyss.
Qwen3.8-Flash-Next GGUFs (BETA). 16G64G = 16 GB GPU + 64 GB RAM; E448 = 448 of 512 experts. Japanese checked; Abyss and llama.cpp.
-
zephel01/Qwen3.8-Flash-Next-E448-IQ2_M-12G32G-GGUF
Text Generation • 162B • Updated • 223 • 5 -
zephel01/Qwen3.8-Flash-Next-IQ3_M-16G64G-GGUF
Text Generation • 177B • Updated • 129 • 1 -
zephel01/Qwen3.8-Flash-Next-IQ3_M-24G64G-GGUF
Text Generation • 177B • Updated • 85 • 1 -
zephel01/Qwen3.8-Flash-Next-DGX-Spark-GGUF
Text Generation • 177B • Updated • 35