Models built for distill — CLI output compression and classification. https://github.com/samuelfaj/distill
Samuel Fajreldines
samuelfaj
AI & ML interests
None yet
Recent Activity
updated a model 8 days ago
samuelfaj/distill-E4B-it-4-bit-MLX published a model 8 days ago
samuelfaj/distill-E4B-it-4-bit-MLX updated a model 10 days ago
samuelfaj/distill3-v0c-datasetOrganizations
None yet
Ornstein 3.6 35B-A3B SABER MLX+MTPLX bundles — fast local serving on Apple Silicon via lightning-mlx. 4/6/8-bit quants.
-
samuelfaj/Ornstein3.6-35B-A3B-SABER-4bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 108 -
samuelfaj/Ornstein3.6-35B-A3B-SABER-6bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 183 • 1 -
samuelfaj/Ornstein3.6-35B-A3B-SABER-8bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 103
Ornstein-Hermes 27B SABER MLX + MTPLX bundles for fast local serving on Apple Silicon via lightning-mlx.
-
samuelfaj/Ornstein-Hermes-3.6-27B-SABER-4bit-MTPLX-Optimized-Speed
Text Generation • 5B • Updated • 629 • 2 -
samuelfaj/Ornstein-Hermes-3.6-27B-SABER-6bit-MTPLX-Optimized-Speed
Text Generation • 6B • Updated • 273 • 1 -
samuelfaj/Ornstein-Hermes-3.6-27B-SABER-8bit-MTPLX-Optimized-Speed
Text Generation • 8B • Updated • 302
Qwopus 3.6 35B-A3B (Qwen3.5 MoE multimodal, reasoning + tool-use) MLX + MTPLX bundles for Apple Silicon via lightning-mlx.
Ornstein 3.6 27B-MTP NSC-ACE-SABER MLX+MTPLX bundles — fast local serving on Apple Silicon via lightning-mlx. 4/6/8-bit quants.
-
samuelfaj/Ornstein3.6-27B-MTP-NSC-ACE-SABER-4bit-MTPLX-Optimized-Speed
Text Generation • 27B • Updated • 234 • 1 -
samuelfaj/Ornstein3.6-27B-MTP-NSC-ACE-SABER-6bit-MTPLX-Optimized-Speed
Text Generation • 27B • Updated • 185 -
samuelfaj/Ornstein3.6-27B-MTP-NSC-ACE-SABER-8bit-MTPLX-Optimized-Speed
Text Generation • 27B • Updated • 380 • 1
QWEN NSC-ACE-SABER MLX + MTPLX bundles for lightning-mlx on Apple Silicon. MTP weights from F16 GGUF source.
-
samuelfaj/Qwen3.6-35B-A3B-NSC-ACE-SABER-4bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 1.01k • 2 -
samuelfaj/Qwen3.6-35B-A3B-NSC-ACE-SABER-6bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 612 -
samuelfaj/Qwen3.6-35B-A3B-NSC-ACE-SABER-8bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 1.61k • 4
Qwen 3.6 35B-A3B (Qwen3.5 MoE) MLX + MTPLX bundles for Apple Silicon via lightning-mlx.
Models built for distill — CLI output compression and classification. https://github.com/samuelfaj/distill
Ornstein 3.6 27B-MTP NSC-ACE-SABER MLX+MTPLX bundles — fast local serving on Apple Silicon via lightning-mlx. 4/6/8-bit quants.
-
samuelfaj/Ornstein3.6-27B-MTP-NSC-ACE-SABER-4bit-MTPLX-Optimized-Speed
Text Generation • 27B • Updated • 234 • 1 -
samuelfaj/Ornstein3.6-27B-MTP-NSC-ACE-SABER-6bit-MTPLX-Optimized-Speed
Text Generation • 27B • Updated • 185 -
samuelfaj/Ornstein3.6-27B-MTP-NSC-ACE-SABER-8bit-MTPLX-Optimized-Speed
Text Generation • 27B • Updated • 380 • 1
Ornstein 3.6 35B-A3B SABER MLX+MTPLX bundles — fast local serving on Apple Silicon via lightning-mlx. 4/6/8-bit quants.
-
samuelfaj/Ornstein3.6-35B-A3B-SABER-4bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 108 -
samuelfaj/Ornstein3.6-35B-A3B-SABER-6bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 183 • 1 -
samuelfaj/Ornstein3.6-35B-A3B-SABER-8bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 103
QWEN NSC-ACE-SABER MLX + MTPLX bundles for lightning-mlx on Apple Silicon. MTP weights from F16 GGUF source.
-
samuelfaj/Qwen3.6-35B-A3B-NSC-ACE-SABER-4bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 1.01k • 2 -
samuelfaj/Qwen3.6-35B-A3B-NSC-ACE-SABER-6bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 612 -
samuelfaj/Qwen3.6-35B-A3B-NSC-ACE-SABER-8bit-MTPLX-Optimized-Speed
Text Generation • 35B • Updated • 1.61k • 4
Ornstein-Hermes 27B SABER MLX + MTPLX bundles for fast local serving on Apple Silicon via lightning-mlx.
-
samuelfaj/Ornstein-Hermes-3.6-27B-SABER-4bit-MTPLX-Optimized-Speed
Text Generation • 5B • Updated • 629 • 2 -
samuelfaj/Ornstein-Hermes-3.6-27B-SABER-6bit-MTPLX-Optimized-Speed
Text Generation • 6B • Updated • 273 • 1 -
samuelfaj/Ornstein-Hermes-3.6-27B-SABER-8bit-MTPLX-Optimized-Speed
Text Generation • 8B • Updated • 302
Qwen 3.6 35B-A3B (Qwen3.5 MoE) MLX + MTPLX bundles for Apple Silicon via lightning-mlx.
Qwopus 3.6 35B-A3B (Qwen3.5 MoE multimodal, reasoning + tool-use) MLX + MTPLX bundles for Apple Silicon via lightning-mlx.