-
warshanks/talkie-1930-13b-it-mlx-4bit-DWQ
Text Generation • 13B • Updated • 20 -
warshanks/talkie-1930-13b-it-mlx-4bit
Text Generation • 13B • Updated • 17 -
warshanks/talkie-1930-13b-it-mlx-6bit
Text Generation • 13B • Updated • 6 -
warshanks/talkie-1930-13b-it-mlx-8bit
Text Generation • 13B • Updated • 14
Ben Shankles PRO
warshanks
AI & ML interests
MLX, AWQ / AI on the edge and in Healthcare
Recent Activity
updated a model about 2 months ago
warshanks/gemma-4-12B-it-assistant-coreai published a model about 2 months ago
warshanks/gemma-4-12B-it-assistant-coreai updated a model about 2 months ago
warshanks/gemma-4-12B-it-coreaiOrganizations
Menlo Research Quantizations
Mistral Quantizations
Lingshu MLX conversions
-
Lingshu: A Generalist Foundation Model for Unified Multimodal Medical Understanding and Reasoning
Paper • 2506.07044 • Published • 114 -
mlx-community/Lingshu-7B-4bit
Image-Text-to-Text • 1B • Updated • 10 -
mlx-community/Lingshu-7B-6bit
Image-Text-to-Text • Updated • 8 -
mlx-community/Lingshu-7B-8bit
Image-Text-to-Text • Updated • 9
Healthcare oriented LLM conversions to MLX
-
mlx-community/medgemma-4b-it-4bit
Image-Text-to-Text • 0.9B • Updated • 1.67k • 3 -
mlx-community/medgemma-4b-it-6bit
Image-Text-to-Text • 1B • Updated • 32 • 1 -
mlx-community/medgemma-4b-it-8bit
Image-Text-to-Text • 1B • Updated • 76 • 3 -
mlx-community/medgemma-4b-it-bf16
Image-Text-to-Text • 5B • Updated • 44 • 1
TheDrummer MLX model conversions I've done
-
warshanks/Qwen3-30B-A3B-Instruct-2507-AWQ
Text Generation • 5B • Updated • 302 • 1 -
warshanks/Qwen3-16B-A3B-abliterated-AWQ
Text Generation • 3B • Updated • 6 • 1 -
warshanks/Huihui-Qwen3-14B-abliterated-v2-AWQ
Text Generation • 3B • Updated • 2.2k • 2 -
warshanks/Qwen3-8B-abliterated-AWQ
Text Generation • 2B • Updated • 1.45k • 1
W4A16 quants made with llmcompressor
MedGemma MLX conversions I've done
-
mlx-community/medgemma-4b-it-4bit
Image-Text-to-Text • 0.9B • Updated • 1.67k • 3 -
mlx-community/medgemma-4b-it-6bit
Image-Text-to-Text • 1B • Updated • 32 • 1 -
mlx-community/medgemma-4b-it-8bit
Image-Text-to-Text • 1B • Updated • 76 • 3 -
mlx-community/medgemma-4b-it-bf16
Image-Text-to-Text • 5B • Updated • 44 • 1
DeepSeek MLX conversions I've done
-
mlx-community/DeepSeek-R1-0528-Qwen3-8B-4bit
Text Generation • 1B • Updated • 1.88k • 6 -
mlx-community/DeepSeek-R1-0528-Qwen3-8B-6bit
Text Generation • 8B • Updated • 24 • 1 -
mlx-community/DeepSeek-R1-0528-Qwen3-8B-8bit
Text Generation • 2B • Updated • 65 • 3 -
mlx-community/DeepSeek-R1-0528-Qwen3-8B-bf16
Text Generation • 8B • Updated • 115 • 4
Nemotron MLX conversions I've done
-
mlx-community/AceReason-Nemotron-7B-4bit
Text Generation • 1B • Updated • 73 -
mlx-community/AceReason-Nemotron-7B-8bit
Text Generation • 2B • Updated • 6 -
mlx-community/AceReason-Nemotron-7B-bf16
Text Generation • 8B • Updated • 5 -
mlx-community/AceReason-Nemotron-1.1-7B-4bit
Text Generation • 1B • Updated • 9
Abliterated model conversions to MLX
-
mlx-community/Josiefied-Health-Qwen3-8B-abliterated-v1-4bit
Text Generation • 1B • Updated • 21 -
mlx-community/Josiefied-Health-Qwen3-8B-abliterated-v1-8bit
Text Generation • 8B • Updated • 30 -
mlx-community/Josiefied-Health-Qwen3-8B-abliterated-v1-bf16
Text Generation • 8B • Updated • 15 • 1 -
mlx-community/Josiefied-DeepSeek-R1-0528-Qwen3-8B-abliterated-v1-4bit
Text Generation • 1B • Updated • 144 • 3
-
warshanks/talkie-1930-13b-it-mlx-4bit-DWQ
Text Generation • 13B • Updated • 20 -
warshanks/talkie-1930-13b-it-mlx-4bit
Text Generation • 13B • Updated • 17 -
warshanks/talkie-1930-13b-it-mlx-6bit
Text Generation • 13B • Updated • 6 -
warshanks/talkie-1930-13b-it-mlx-8bit
Text Generation • 13B • Updated • 14
-
warshanks/Qwen3-30B-A3B-Instruct-2507-AWQ
Text Generation • 5B • Updated • 302 • 1 -
warshanks/Qwen3-16B-A3B-abliterated-AWQ
Text Generation • 3B • Updated • 6 • 1 -
warshanks/Huihui-Qwen3-14B-abliterated-v2-AWQ
Text Generation • 3B • Updated • 2.2k • 2 -
warshanks/Qwen3-8B-abliterated-AWQ
Text Generation • 2B • Updated • 1.45k • 1
Menlo Research Quantizations
W4A16 quants made with llmcompressor
Mistral Quantizations
MedGemma MLX conversions I've done
-
mlx-community/medgemma-4b-it-4bit
Image-Text-to-Text • 0.9B • Updated • 1.67k • 3 -
mlx-community/medgemma-4b-it-6bit
Image-Text-to-Text • 1B • Updated • 32 • 1 -
mlx-community/medgemma-4b-it-8bit
Image-Text-to-Text • 1B • Updated • 76 • 3 -
mlx-community/medgemma-4b-it-bf16
Image-Text-to-Text • 5B • Updated • 44 • 1
Lingshu MLX conversions
-
Lingshu: A Generalist Foundation Model for Unified Multimodal Medical Understanding and Reasoning
Paper • 2506.07044 • Published • 114 -
mlx-community/Lingshu-7B-4bit
Image-Text-to-Text • 1B • Updated • 10 -
mlx-community/Lingshu-7B-6bit
Image-Text-to-Text • Updated • 8 -
mlx-community/Lingshu-7B-8bit
Image-Text-to-Text • Updated • 9
DeepSeek MLX conversions I've done
-
mlx-community/DeepSeek-R1-0528-Qwen3-8B-4bit
Text Generation • 1B • Updated • 1.88k • 6 -
mlx-community/DeepSeek-R1-0528-Qwen3-8B-6bit
Text Generation • 8B • Updated • 24 • 1 -
mlx-community/DeepSeek-R1-0528-Qwen3-8B-8bit
Text Generation • 2B • Updated • 65 • 3 -
mlx-community/DeepSeek-R1-0528-Qwen3-8B-bf16
Text Generation • 8B • Updated • 115 • 4
Healthcare oriented LLM conversions to MLX
-
mlx-community/medgemma-4b-it-4bit
Image-Text-to-Text • 0.9B • Updated • 1.67k • 3 -
mlx-community/medgemma-4b-it-6bit
Image-Text-to-Text • 1B • Updated • 32 • 1 -
mlx-community/medgemma-4b-it-8bit
Image-Text-to-Text • 1B • Updated • 76 • 3 -
mlx-community/medgemma-4b-it-bf16
Image-Text-to-Text • 5B • Updated • 44 • 1
Nemotron MLX conversions I've done
-
mlx-community/AceReason-Nemotron-7B-4bit
Text Generation • 1B • Updated • 73 -
mlx-community/AceReason-Nemotron-7B-8bit
Text Generation • 2B • Updated • 6 -
mlx-community/AceReason-Nemotron-7B-bf16
Text Generation • 8B • Updated • 5 -
mlx-community/AceReason-Nemotron-1.1-7B-4bit
Text Generation • 1B • Updated • 9
TheDrummer MLX model conversions I've done
Abliterated model conversions to MLX
-
mlx-community/Josiefied-Health-Qwen3-8B-abliterated-v1-4bit
Text Generation • 1B • Updated • 21 -
mlx-community/Josiefied-Health-Qwen3-8B-abliterated-v1-8bit
Text Generation • 8B • Updated • 30 -
mlx-community/Josiefied-Health-Qwen3-8B-abliterated-v1-bf16
Text Generation • 8B • Updated • 15 • 1 -
mlx-community/Josiefied-DeepSeek-R1-0528-Qwen3-8B-abliterated-v1-4bit
Text Generation • 1B • Updated • 144 • 3