Quants created recently.. where time is relative
Gheorghe Chesler PRO
nightmedia
AI & ML interests
Nightmedia: human-Like AI and the MLX Deckard(qx) Formula
Donations are appreciated:
BTC:36d7U1n3MFaXgnNRAaEL3Pa3Hy6oFhM7XY
Recent Activity
updated
a model
about 7 hours ago
nightmedia/Gemma-3-27b-it-HERETIC-Gemini-1000x-Deep-Reasoning-q6-mlx
updated
a collection
about 16 hours ago
Runs On My 32GB Mac
updated
a collection
about 16 hours ago
Runs On My 32GB Mac
Organizations
Models used in Nightmedia merges
An image is worth a thousand vibes
-
nightmedia/Qwen3-VL-30B-A3B-Instruct-qx86-hi-mlx
Image-Text-to-Text • Updated • 12 • 1 -
nightmedia/Kimi-VL-A3B-Thinking-2506-q8-hi-mlx
Text Generation • 16B • Updated • 29 • 1 -
nightmedia/Qwen3-VL-30B-A3B-Thinking-qx64-hi-mlx
Text Generation • Updated • 8 -
nightmedia/Qwen3-VL-30B-A3B-Instruct-qx64-hi-mlx
Image-Text-to-Text • Updated • 13
some slower than others
-
nightmedia/Mistral-Devstral-2505-CODER-Brainstorm40x-44B-q6-mlx
Text Generation • 44B • Updated • 25 -
nightmedia/Mistral-Devstral-2507-CODER-Brainstorm40x-44B-q6-mlx
Text Generation • 44B • Updated • 32 -
nightmedia/KAT-Dev-72B-Exp-qx86-hi-mlx
Text Generation • 73B • Updated • 8 -
nightmedia/KAT-Dev-qx86-hi-mlx
Text Generation • 33B • Updated • 6 • 1
small, fast, different, a bit dry
Brainstormed models by DavidAU
-
nightmedia/Qwen3-Deckard-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 3 • 1 -
nightmedia/Qwen3-Deckard-Large-6B-qx86-hi-mlx
6B • Updated • 3 • 1 -
nightmedia/Qwen3-Deckard-Large-Heavy-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 1 -
nightmedia/Qwen3-Deckard-Large-Almost-Human-6B-qx86-hi-mlx
Text Generation • 6B • Updated
Emergent behavior
-
nightmedia/Qwen3-53B-A3B-TOTAL-RECALL-MASTER-CODER-v1.4-q6-mlx
Text Generation • 53B • Updated • 92 • 1 -
nightmedia/Qwen3-42B-A3B-2507-Thinking-Abliterated-uncensored-TOTAL-RECALL-v2-Medium-MASTER-CODER-qx6-mlx
Text Generation • 42B • Updated • 94 -
nightmedia/Qwen3-Darkest-BIG-Jan-Horror-v1-256k-ctx-8B-qx86-hi-mlx
Text Generation • 8B • Updated • 10 -
nightmedia/unsloth-Qwen3-Next-80B-A3B-Instruct-1M-qx65-hi-mlx
Text Generation • 80B • Updated • 57 • 1
Trained to be different
-
nightmedia/Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B-q6-f32-mlx
Text Generation • 24B • Updated • 78 -
nightmedia/LLama-3.1-128k-Darkest-Planet-Uncensored-16.5B-q6-mlx
Text Generation • 17B • Updated • 31 -
nightmedia/LLama-3.1-128k-Darkest-Planet-Uncensored-16.5B-q8-mlx
Text Generation • 17B • Updated • 26 -
nightmedia/LLama-3.1-128k-Darkest-Planet-Uncensored-16.5B-q5-mlx
Text Generation • 17B • Updated • 28
Coding with Spock
-
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-ST-TNG-qx64-hi-mlx
Text Generation • 42B • Updated • 1 -
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-ST-TNG-qx86-hi-mlx
Text Generation • 42B • Updated • 2 -
nightmedia/Qwen3-ST-TNG-DS9-Voyager-Unhinged-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 18 • 1 -
nightmedia/Qwen3-ST-TNG-DS9-Voyager-Unhinged2-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 13
The world is your oyster—for small pearls
-
nightmedia/Qwen3-53B-A3B-TOTAL-RECALL-MASTER-CODER-v1.4-q6-mlx
Text Generation • 53B • Updated • 92 • 1 -
nightmedia/Qwen3-30B-A3B-Thinking-2507-q5-mlx
Text Generation • 31B • Updated • 22 • 1 -
nightmedia/Qwen3-30B-A3B-YOYO-V2-qx6-mxfp4-mlx
Text Generation • 31B • Updated • 13 • 1 -
nightmedia/Qwen3-30B-A3B-YOYO-V3-qx86-hi-mlx
Text Generation • 31B • Updated • 26 • 2
Some with full context
-
nightmedia/Qwen3-4B-BiasExpert-dwq6-mlx
Text Generation • 1B • Updated • 14 • 2 -
nightmedia/Qwen3-ST-TNG-DS9-Voyager-Unhinged-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 18 • 1 -
nightmedia/Qwen3-Deckard-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 3 • 1 -
nightmedia/Qwen3-4B-Agent-F32-dwq4-mlx
Text Generation • 0.8B • Updated • 200 • 2
Quants that make a difference
-
nightmedia/Qwen3-Next-80B-A3B-Thinking-1M-qx86n-hi-mlx
Text Generation • 80B • Updated • 78 • 3 -
nightmedia/Qwen3-Next-80B-A3B-Instruct-qx86n-mlx
Text Generation • 80B • Updated • 28 • 2 -
nightmedia/Qwen3-Next-80B-A3B-Instruct-qx64-mlx
Text Generation • 80B • Updated • 48 • 4 -
nightmedia/unsloth-GLM-4.5-Air-qx64-mlx
Text Generation • 107B • Updated • 34 • 1
The little engines that could
-
nightmedia/Qwen3-Deckard-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 3 • 1 -
nightmedia/Qwen3-VLTO-1.7B-Instruct-qx85x-mlx
Text Generation • 0.4B • Updated • 10 -
nightmedia/Qwen3-VLTO-1.7B-Instruct-qx85x-hi-mlx
Text Generation • 0.4B • Updated • 13 -
nightmedia/Qwen3-4B-Agent-F32-dwq4-mlx
Text Generation • 0.8B • Updated • 200 • 2
Unleash the inner experts
-
nightmedia/Qwen3-30B-A3B-YOYO-V3-qx86-hi-mlx
Text Generation • 31B • Updated • 26 • 2 -
nightmedia/unsloth-Qwen3-Next-80B-A3B-Instruct-qx86-hi-mlx
Text Generation • 80B • Updated • 83 • 2 -
nightmedia/Qwen3-Next-80B-A3B-Thinking-1M-qx86-hi-mlx
Text Generation • 80B • Updated • 21 -
nightmedia/Qwen3-Next-80B-A3B-Thinking-qx64-hi-mlx
Text Generation • 80B • Updated • 26 • 1
Two heads(or more) are better than one
-
nightmedia/Qwen3-Esper3-Reasoning-CODER-Instruct-21B-Brainstorm20x-128k-ctx-q6-mlx
Text Generation • 21B • Updated • 10 -
nightmedia/Qwen3-MOE-6x6B-Star-Trek-Universe-Alpha-D-256k-ctx-36B-qx65-mlx
Text Generation • 27B • Updated • 15 • 1 -
nightmedia/Qwen3-2x6B-TNG-Deckard-Alpha-III-12B-mlx
Text Generation • 10B • Updated • 1 -
nightmedia/Qwen3-2x6B-TNG-Deckard-Alpha-12B-LH-mlx
Text Generation • 10B • Updated • 2
The doctor is in
Models with depth
-
nightmedia/Qwen3-53B-A3B-TOTAL-RECALL-MASTER-CODER-v1.4-q6-mlx
Text Generation • 53B • Updated • 92 • 1 -
nightmedia/Qwen3-30B-A3B-YOYO-V2-qx6-mxfp4-mlx
Text Generation • 31B • Updated • 13 • 1 -
nightmedia/Qwen3-53B-A3B-2507-THINKING-TOTAL-RECALL-v2-MASTER-CODER-1m-dwq4-mlx
Text Generation • 53B • Updated • 100 • 1 -
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-Total-Recall-qx64-hi-mlx
Text Generation • 42B • Updated • 5 • 1
brainstorming MoEs by DavidAU
-
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-ST-TNG-qx64-hi-mlx
Text Generation • 42B • Updated • 1 -
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-Horror-qx86-hi-mlx
Text Generation • 42B • Updated • 7 -
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-ST-TNG-qx86-hi-mlx
Text Generation • 42B • Updated • 2 -
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-ST-TNG-II-qx86-hi-mlx
Text Generation • 42B • Updated • 1
Deckard, the personal assistant
-
nightmedia/Qwen3-Deckard-Large-Almost-Human-6B-qx86-hi-mlx
Text Generation • 6B • Updated -
nightmedia/Qwen3-Deckard-Large-Almost-Human-6B-II-qx86-hi-mlx
Text Generation • 6B • Updated • 10 -
nightmedia/Llama-3.2-PKD-Deckard-Almost-Human-abliterated-uncensored-7B-qx86-hi-mlx
Text Generation • 7B • Updated • 229 • 1 -
nightmedia/Qwen3-Deckard-Large-Heavy-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 1
96GB and up
-
nightmedia/GLM-4.5-Air-q3-hi-mlx
Text Generation • 107B • Updated • 15 • 1 -
nightmedia/unsloth-gpt-oss-120b-q5-hi-mlx
Text Generation • 117B • Updated • 50 -
nightmedia/unsloth-GLM-4.5-Air-qx64-mlx
Text Generation • 107B • Updated • 34 • 1 -
nightmedia/unsloth-GLM-4.5-Air-mxfp4-mlx
Text Generation • 107B • Updated • 39 • 1
Some smarter, and some MoEs
-
nightmedia/Qwen3-30B-A3B-Thinking-2507-q5-mlx
Text Generation • 31B • Updated • 22 • 1 -
nightmedia/Qwen3-Esper3-Reasoning-CODER-Instruct-21B-Brainstorm20x-128k-ctx-q6-mlx
Text Generation • 21B • Updated • 10 -
nightmedia/Qwen3-MOE-6x6B-Star-Trek-Universe-Alpha-D-256k-ctx-36B-qx65-mlx
Text Generation • 27B • Updated • 15 • 1 -
nightmedia/Qwen3-2x6B-TNG-Deckard-Alpha-III-12B-mlx
Text Generation • 10B • Updated • 1
All I got.. but eager to code
Quants created recently.. where time is relative
Quants that make a difference
-
nightmedia/Qwen3-Next-80B-A3B-Thinking-1M-qx86n-hi-mlx
Text Generation • 80B • Updated • 78 • 3 -
nightmedia/Qwen3-Next-80B-A3B-Instruct-qx86n-mlx
Text Generation • 80B • Updated • 28 • 2 -
nightmedia/Qwen3-Next-80B-A3B-Instruct-qx64-mlx
Text Generation • 80B • Updated • 48 • 4 -
nightmedia/unsloth-GLM-4.5-Air-qx64-mlx
Text Generation • 107B • Updated • 34 • 1
Models used in Nightmedia merges
The little engines that could
-
nightmedia/Qwen3-Deckard-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 3 • 1 -
nightmedia/Qwen3-VLTO-1.7B-Instruct-qx85x-mlx
Text Generation • 0.4B • Updated • 10 -
nightmedia/Qwen3-VLTO-1.7B-Instruct-qx85x-hi-mlx
Text Generation • 0.4B • Updated • 13 -
nightmedia/Qwen3-4B-Agent-F32-dwq4-mlx
Text Generation • 0.8B • Updated • 200 • 2
An image is worth a thousand vibes
-
nightmedia/Qwen3-VL-30B-A3B-Instruct-qx86-hi-mlx
Image-Text-to-Text • Updated • 12 • 1 -
nightmedia/Kimi-VL-A3B-Thinking-2506-q8-hi-mlx
Text Generation • 16B • Updated • 29 • 1 -
nightmedia/Qwen3-VL-30B-A3B-Thinking-qx64-hi-mlx
Text Generation • Updated • 8 -
nightmedia/Qwen3-VL-30B-A3B-Instruct-qx64-hi-mlx
Image-Text-to-Text • Updated • 13
Unleash the inner experts
-
nightmedia/Qwen3-30B-A3B-YOYO-V3-qx86-hi-mlx
Text Generation • 31B • Updated • 26 • 2 -
nightmedia/unsloth-Qwen3-Next-80B-A3B-Instruct-qx86-hi-mlx
Text Generation • 80B • Updated • 83 • 2 -
nightmedia/Qwen3-Next-80B-A3B-Thinking-1M-qx86-hi-mlx
Text Generation • 80B • Updated • 21 -
nightmedia/Qwen3-Next-80B-A3B-Thinking-qx64-hi-mlx
Text Generation • 80B • Updated • 26 • 1
some slower than others
-
nightmedia/Mistral-Devstral-2505-CODER-Brainstorm40x-44B-q6-mlx
Text Generation • 44B • Updated • 25 -
nightmedia/Mistral-Devstral-2507-CODER-Brainstorm40x-44B-q6-mlx
Text Generation • 44B • Updated • 32 -
nightmedia/KAT-Dev-72B-Exp-qx86-hi-mlx
Text Generation • 73B • Updated • 8 -
nightmedia/KAT-Dev-qx86-hi-mlx
Text Generation • 33B • Updated • 6 • 1
Two heads(or more) are better than one
-
nightmedia/Qwen3-Esper3-Reasoning-CODER-Instruct-21B-Brainstorm20x-128k-ctx-q6-mlx
Text Generation • 21B • Updated • 10 -
nightmedia/Qwen3-MOE-6x6B-Star-Trek-Universe-Alpha-D-256k-ctx-36B-qx65-mlx
Text Generation • 27B • Updated • 15 • 1 -
nightmedia/Qwen3-2x6B-TNG-Deckard-Alpha-III-12B-mlx
Text Generation • 10B • Updated • 1 -
nightmedia/Qwen3-2x6B-TNG-Deckard-Alpha-12B-LH-mlx
Text Generation • 10B • Updated • 2
small, fast, different, a bit dry
The doctor is in
Brainstormed models by DavidAU
-
nightmedia/Qwen3-Deckard-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 3 • 1 -
nightmedia/Qwen3-Deckard-Large-6B-qx86-hi-mlx
6B • Updated • 3 • 1 -
nightmedia/Qwen3-Deckard-Large-Heavy-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 1 -
nightmedia/Qwen3-Deckard-Large-Almost-Human-6B-qx86-hi-mlx
Text Generation • 6B • Updated
Models with depth
-
nightmedia/Qwen3-53B-A3B-TOTAL-RECALL-MASTER-CODER-v1.4-q6-mlx
Text Generation • 53B • Updated • 92 • 1 -
nightmedia/Qwen3-30B-A3B-YOYO-V2-qx6-mxfp4-mlx
Text Generation • 31B • Updated • 13 • 1 -
nightmedia/Qwen3-53B-A3B-2507-THINKING-TOTAL-RECALL-v2-MASTER-CODER-1m-dwq4-mlx
Text Generation • 53B • Updated • 100 • 1 -
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-Total-Recall-qx64-hi-mlx
Text Generation • 42B • Updated • 5 • 1
Emergent behavior
-
nightmedia/Qwen3-53B-A3B-TOTAL-RECALL-MASTER-CODER-v1.4-q6-mlx
Text Generation • 53B • Updated • 92 • 1 -
nightmedia/Qwen3-42B-A3B-2507-Thinking-Abliterated-uncensored-TOTAL-RECALL-v2-Medium-MASTER-CODER-qx6-mlx
Text Generation • 42B • Updated • 94 -
nightmedia/Qwen3-Darkest-BIG-Jan-Horror-v1-256k-ctx-8B-qx86-hi-mlx
Text Generation • 8B • Updated • 10 -
nightmedia/unsloth-Qwen3-Next-80B-A3B-Instruct-1M-qx65-hi-mlx
Text Generation • 80B • Updated • 57 • 1
brainstorming MoEs by DavidAU
-
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-ST-TNG-qx64-hi-mlx
Text Generation • 42B • Updated • 1 -
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-Horror-qx86-hi-mlx
Text Generation • 42B • Updated • 7 -
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-ST-TNG-qx86-hi-mlx
Text Generation • 42B • Updated • 2 -
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-ST-TNG-II-qx86-hi-mlx
Text Generation • 42B • Updated • 1
Trained to be different
-
nightmedia/Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B-q6-f32-mlx
Text Generation • 24B • Updated • 78 -
nightmedia/LLama-3.1-128k-Darkest-Planet-Uncensored-16.5B-q6-mlx
Text Generation • 17B • Updated • 31 -
nightmedia/LLama-3.1-128k-Darkest-Planet-Uncensored-16.5B-q8-mlx
Text Generation • 17B • Updated • 26 -
nightmedia/LLama-3.1-128k-Darkest-Planet-Uncensored-16.5B-q5-mlx
Text Generation • 17B • Updated • 28
Deckard, the personal assistant
-
nightmedia/Qwen3-Deckard-Large-Almost-Human-6B-qx86-hi-mlx
Text Generation • 6B • Updated -
nightmedia/Qwen3-Deckard-Large-Almost-Human-6B-II-qx86-hi-mlx
Text Generation • 6B • Updated • 10 -
nightmedia/Llama-3.2-PKD-Deckard-Almost-Human-abliterated-uncensored-7B-qx86-hi-mlx
Text Generation • 7B • Updated • 229 • 1 -
nightmedia/Qwen3-Deckard-Large-Heavy-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 1
Coding with Spock
-
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-ST-TNG-qx64-hi-mlx
Text Generation • 42B • Updated • 1 -
nightmedia/Qwen3-Yoyo-V3-42B-A3B-Thinking-TOTAL-RECALL-ST-TNG-qx86-hi-mlx
Text Generation • 42B • Updated • 2 -
nightmedia/Qwen3-ST-TNG-DS9-Voyager-Unhinged-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 18 • 1 -
nightmedia/Qwen3-ST-TNG-DS9-Voyager-Unhinged2-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 13
96GB and up
-
nightmedia/GLM-4.5-Air-q3-hi-mlx
Text Generation • 107B • Updated • 15 • 1 -
nightmedia/unsloth-gpt-oss-120b-q5-hi-mlx
Text Generation • 117B • Updated • 50 -
nightmedia/unsloth-GLM-4.5-Air-qx64-mlx
Text Generation • 107B • Updated • 34 • 1 -
nightmedia/unsloth-GLM-4.5-Air-mxfp4-mlx
Text Generation • 107B • Updated • 39 • 1
The world is your oyster—for small pearls
-
nightmedia/Qwen3-53B-A3B-TOTAL-RECALL-MASTER-CODER-v1.4-q6-mlx
Text Generation • 53B • Updated • 92 • 1 -
nightmedia/Qwen3-30B-A3B-Thinking-2507-q5-mlx
Text Generation • 31B • Updated • 22 • 1 -
nightmedia/Qwen3-30B-A3B-YOYO-V2-qx6-mxfp4-mlx
Text Generation • 31B • Updated • 13 • 1 -
nightmedia/Qwen3-30B-A3B-YOYO-V3-qx86-hi-mlx
Text Generation • 31B • Updated • 26 • 2
Some smarter, and some MoEs
-
nightmedia/Qwen3-30B-A3B-Thinking-2507-q5-mlx
Text Generation • 31B • Updated • 22 • 1 -
nightmedia/Qwen3-Esper3-Reasoning-CODER-Instruct-21B-Brainstorm20x-128k-ctx-q6-mlx
Text Generation • 21B • Updated • 10 -
nightmedia/Qwen3-MOE-6x6B-Star-Trek-Universe-Alpha-D-256k-ctx-36B-qx65-mlx
Text Generation • 27B • Updated • 15 • 1 -
nightmedia/Qwen3-2x6B-TNG-Deckard-Alpha-III-12B-mlx
Text Generation • 10B • Updated • 1
Some with full context
-
nightmedia/Qwen3-4B-BiasExpert-dwq6-mlx
Text Generation • 1B • Updated • 14 • 2 -
nightmedia/Qwen3-ST-TNG-DS9-Voyager-Unhinged-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 18 • 1 -
nightmedia/Qwen3-Deckard-6B-qx86-hi-mlx
Text Generation • 6B • Updated • 3 • 1 -
nightmedia/Qwen3-4B-Agent-F32-dwq4-mlx
Text Generation • 0.8B • Updated • 200 • 2
All I got.. but eager to code