Gemma AXQ research packs. Model-specific runtime evidence; no inherited quality or MTP-speed certification.
AutomatosX PRO
AutomatosX
AI & ML interests
None yet
Recent Activity
updated a model 4 days ago
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP8-MTP updated a model 4 days ago
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP4-MTP updated a model 4 days ago
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-6bit-MTPOrganizations
MLX affine6 base packs with precision protection. MTP where packaged; see measured BPW and per-model runtime evidence. Not certified.
-
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-6bit-MTP
Text Generation • 35B • Updated • 175 -
AutomatosX/AX-Cyber-Tiel-Coder-35B-A3B-MLX-AXQ-6bit-MTP
Text Generation • 35B • Updated • 251 -
AutomatosX/AX-Qwen3.8-Flash-Next-MLX-AXQ-6bit-MTP
Text Generation • 177B • Updated • 81 -
AutomatosX/AX-Swift-1.5-Qwen3.8-27B-MLX-AXQ-6bit-MTP
Text Generation • 27B • Updated • 216
MLX AXQ MXFP8 development checkpoints; protected tensors keep their declared precision.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93 -
AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-MXFP8
Image-Text-to-Text • 3B • Updated • 144 -
AutomatosX/AX-Unlimited-OCR-3B-MoE-MLX-AXQ-MXFP8
Image-Text-to-Text • 3B • Updated • 264
AXQ packs with native MTP weights or exact-pair external assistants. Packaging, execution, acceptance and speed are separate evidence.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93
Low-precision reference storage, including source-layout MX matrices. See reconstruction and runtime limits in each card; not quality-certified.
AX BI selected AXQ Qwen and Nemotron MLX MXFP4/MXFP8 packs with protected MTP. Development artifacts; see individual model cards.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93
Qwen3.8, Qwen3.6, Qwen3.5, VL, embeddings, and image reference storage.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93
Nemotron-3-Embed 1B/8B retrieval and Super-120B/Lightning-30B MTP AXQ packs: MLX MXFP4/MXFP8 and CUDA NVFP4. Not certified.
-
AutomatosX/AX-Nemotron-3-Embed-1B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 1B • Updated • 38 -
AutomatosX/AX-Nemotron-3-Embed-8B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 8B • Updated • 40 -
AutomatosX/AX-Nemotron-3-Super-120B-A12B-CUDA-AXQ-NVFP4-MTP
Text Generation • 124B • Updated • 326 -
AutomatosX/AX-Nemotron-3.5-Lightning-30B-A3B-CUDA-AXQ-NVFP4-MTP
Text Generation • 33B • Updated • 384
Tiel Coder 35B-A3B affine6/MXFP4/MXFP8 with packaged MTP, and the Cyber variant.
-
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 751 • 1 -
AutomatosX/AX-Cyber-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 797 -
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 236 -
AutomatosX/AX-Cyber-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 323
Official 35B-A3B MiniCPM-V 4.7 AXQ MLX MXFP4/MXFP8 and NVFP4 development packs. CPU-reference NVFP4; no certification.
-
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-MLX-AXQ-MXFP4
Image-Text-to-Text • 35B • Updated • 121 -
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-MLX-AXQ-MXFP8
Image-Text-to-Text • 35B • Updated • 158 -
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 35B • Updated • 36 -
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-MLX-AXQ-6bit
Image-Text-to-Text • 35B • Updated • 111
Cloudflare Clef, Clef-Flash, and Clef-Omni: MLX MXFP4/MXFP8 and CUDA NVFP4 W4A16.
-
AutomatosX/AX-Clef-Flash-9B-CUDA-AXQ-NVFP4-W4A16
Text Generation • 9B • Updated • 36 -
AutomatosX/AX-Clef-27B-CUDA-AXQ-NVFP4-W4A16
Text Generation • 27B • Updated • 43 -
AutomatosX/AX-Clef-Omni-30B-A3B-CUDA-AXQ-NVFP4-W4A16
Text Generation • 35B • Updated • 38 -
AutomatosX/AX-Clef-Flash-9B-MLX-AXQ-MXFP4
Text Generation • 9B • Updated • 54
DeepSeek-OCR-2 and Unlimited-OCR v1/v2 development packs. Runtime evidence is pack-specific; see each model card.
-
AutomatosX/AX-DeepSeek-OCR-2-CUDA-AXQ-NVFP4-W4A4
Image-Text-to-Text • 3B • Updated • 51 -
AutomatosX/AX-Unlimited-OCR-3B-MoE-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 3B • Updated • 51 -
AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-MXFP8
Image-Text-to-Text • 3B • Updated • 144 -
AutomatosX/AX-DeepSeek-OCR-2-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 3B • Updated • 54
Hy-MT2 MLX MXFP4/MXFP8 and CUDA NVFP4 W4A16. Not certified.
-
AutomatosX/AX-Hy-MT2-30B-A3B-CUDA-AXQ-NVFP4-W4A16
Translation • 30B • Updated • 66 -
AutomatosX/AX-Hy-MT2-30B-A3B-MLX-AXQ-MXFP4
Text Generation • 30B • Updated • 277 -
AutomatosX/AX-Hy-MT2-30B-A3B-MLX-AXQ-MXFP8
Text Generation • 30B • Updated • 99 -
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-MLX-AXQ-MXFP4
Image-Text-to-Text • 35B • Updated • 121
Ornith 9B and 35B AXQ research packs. Ornith 397B is withdrawn. See per-model runtime and MTP evidence; not certified.
MLX AXQ MXFP4 development checkpoints; reference-only storage is listed separately.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.8-27B-MLX-AXQ-MXFP4-MTP
Text Generation • 27B • Updated • 853 -
AutomatosX/AX-Swift-1.5-Qwen3.8-27B-MLX-AXQ-MXFP4-MTP
Text Generation • 27B • Updated • 577 • 1
NVFP4 W4A16/W4A4 development packs; CPU-reference exports are identified per card. CUDA runtime evidence is model-specific.
-
AutomatosX/AX-Qwen3.8-27B-CUDA-AXQ-NVFP4-W4A16-MTP
Image-Text-to-Text • 28B • Updated • 106 -
AutomatosX/AX-Qwen3.5-9B-CUDA-AXQ-NVFP4-W4A16-MTP
Image-Text-to-Text • 10B • Updated • 46 -
AutomatosX/AX-Muse-Glimmer-30B-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 30B • Updated • 65 -
AutomatosX/AX-Qwen3-Embedding-0.6B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 0.6B • Updated • 44
NVFP4 development packs, including CPU-reference exports, and AXDocLayout FP32 CUDA TF32 sources. Runtime evidence is model-specific.
-
AutomatosX/AXDocLayout-M-v1.3-CUDA-TF32
Object Detection • 36.7M • Updated • 36 -
AutomatosX/AXDocLayout-M-v1.2-CUDA-TF32
Object Detection • 36.7M • Updated • 134 -
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 35B • Updated • 36 -
AutomatosX/AX-Qwen3.8-27B-CUDA-AXQ-NVFP4-W4A16-MTP
Image-Text-to-Text • 28B • Updated • 106
Muse Glimmer NVFP4 W4A16 development conversion. Vision stays at source precision. Not quality-certified.
AX Work selected AXQ Qwen and Nemotron MLX MXFP4/MXFP8 packs with protected MTP. Development artifacts; see individual model cards.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93
AXQ embedding packs: MLX affine6/MXFP4/MXFP8 and CUDA NVFP4. See task-specific loader, pooling and runtime evidence. Not certified.
-
AutomatosX/AX-Qwen3-Embedding-0.6B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 0.6B • Updated • 44 -
AutomatosX/AX-Qwen3-Embedding-4B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 4B • Updated • 44 -
AutomatosX/AX-Qwen3-Embedding-8B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 8B • Updated • 45 -
AutomatosX/AX-Nemotron-3-Embed-1B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 1B • Updated • 38
Qwen3.8: MLX affine6/MXFP4/MXFP8 and CUDA NVFP4 development packs with packaged MTP.
-
AutomatosX/AX-Qwen3.8-27B-CUDA-AXQ-NVFP4-W4A16-MTP
Image-Text-to-Text • 28B • Updated • 106 -
AutomatosX/AX-Qwen3.8-27B-MLX-AXQ-MXFP4-MTP
Text Generation • 27B • Updated • 853 -
AutomatosX/AX-Swift-1.5-Qwen3.8-27B-MLX-AXQ-MXFP4-MTP
Text Generation • 27B • Updated • 577 • 1 -
AutomatosX/AX-Qwen3.8-Flash-Next-MLX-AXQ-MXFP4-MTP
Text Generation • 177B • Updated • 599
Coding-source AXQ packs including Tiel, Cyber, Swift, Flash Next and Qwen3-Coder-Next. Check per-model runtime and MTP evidence.
-
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 751 • 1 -
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 236 -
AutomatosX/AX-Swift-1.5-Qwen3.8-27B-MLX-AXQ-MXFP4-MTP
Text Generation • 27B • Updated • 577 • 1 -
AutomatosX/AX-Swift-1.5-Qwen3.8-27B-MLX-AXQ-MXFP8-MTP
Text Generation • 27B • Updated • 329
Qwen3-VL MLX affine6/MXFP4/MXFP8 and CUDA NVFP4 packs. Protected vision; model-specific image smoke evidence. Not certified.
-
AutomatosX/AX-Qwen3-VL-4B-Instruct-CUDA-AXQ-NVFP4-W4A4
Image-Text-to-Text • 4B • Updated • 38 -
AutomatosX/AX-Qwen3-VL-8B-Instruct-CUDA-AXQ-NVFP4-W4A4
Image-Text-to-Text • 9B • Updated • 37 -
AutomatosX/AX-Qwen3-VL-4B-Instruct-MLX-AXQ-MXFP4
Image-Text-to-Text • 4B • Updated • 70 -
AutomatosX/AX-Qwen3-VL-4B-Instruct-MLX-AXQ-MXFP8
Image-Text-to-Text • 4B • Updated • 67
DeepSeek-OCR-2 MLX MXFP8 and CUDA NVFP4 W4A16/W4A4 development packs.
-
AutomatosX/AX-DeepSeek-OCR-2-CUDA-AXQ-NVFP4-W4A4
Image-Text-to-Text • 3B • Updated • 51 -
AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-MXFP8
Image-Text-to-Text • 3B • Updated • 144 -
AutomatosX/AX-DeepSeek-OCR-2-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 3B • Updated • 54 -
AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-6bit
Image-Text-to-Text • 3B • Updated • 99
Unlimited-OCR v1 development packs and v2 MXFP4/MXFP8/NVFP4 reference storage. Runtime evidence is pack-specific.
-
AutomatosX/AX-Unlimited-OCR-3B-MoE-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 3B • Updated • 51 -
AutomatosX/AX-Unlimited-OCR-3B-MoE-MLX-AXQ-MXFP8
Image-Text-to-Text • 3B • Updated • 264 -
AutomatosX/AX-Unlimited-OCR-3B-MoE-MLX-AXQ-6bit
Image-Text-to-Text • 3B • Updated • 138 -
AutomatosX/AX-Unlimited-OCR-2-CUDA-AXQ-NVFP4-W4A16
3B • Updated • 55 • 1
MLX packs only, including model-specific detectors. Native NVIDIA checkpoints are listed in the CUDA collection.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93
Gemma AXQ research packs. Model-specific runtime evidence; no inherited quality or MTP-speed certification.
Ornith 9B and 35B AXQ research packs. Ornith 397B is withdrawn. See per-model runtime and MTP evidence; not certified.
MLX affine6 base packs with precision protection. MTP where packaged; see measured BPW and per-model runtime evidence. Not certified.
-
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-6bit-MTP
Text Generation • 35B • Updated • 175 -
AutomatosX/AX-Cyber-Tiel-Coder-35B-A3B-MLX-AXQ-6bit-MTP
Text Generation • 35B • Updated • 251 -
AutomatosX/AX-Qwen3.8-Flash-Next-MLX-AXQ-6bit-MTP
Text Generation • 177B • Updated • 81 -
AutomatosX/AX-Swift-1.5-Qwen3.8-27B-MLX-AXQ-6bit-MTP
Text Generation • 27B • Updated • 216
MLX AXQ MXFP4 development checkpoints; reference-only storage is listed separately.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.8-27B-MLX-AXQ-MXFP4-MTP
Text Generation • 27B • Updated • 853 -
AutomatosX/AX-Swift-1.5-Qwen3.8-27B-MLX-AXQ-MXFP4-MTP
Text Generation • 27B • Updated • 577 • 1
MLX AXQ MXFP8 development checkpoints; protected tensors keep their declared precision.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93 -
AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-MXFP8
Image-Text-to-Text • 3B • Updated • 144 -
AutomatosX/AX-Unlimited-OCR-3B-MoE-MLX-AXQ-MXFP8
Image-Text-to-Text • 3B • Updated • 264
NVFP4 W4A16/W4A4 development packs; CPU-reference exports are identified per card. CUDA runtime evidence is model-specific.
-
AutomatosX/AX-Qwen3.8-27B-CUDA-AXQ-NVFP4-W4A16-MTP
Image-Text-to-Text • 28B • Updated • 106 -
AutomatosX/AX-Qwen3.5-9B-CUDA-AXQ-NVFP4-W4A16-MTP
Image-Text-to-Text • 10B • Updated • 46 -
AutomatosX/AX-Muse-Glimmer-30B-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 30B • Updated • 65 -
AutomatosX/AX-Qwen3-Embedding-0.6B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 0.6B • Updated • 44
AXQ packs with native MTP weights or exact-pair external assistants. Packaging, execution, acceptance and speed are separate evidence.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93
NVFP4 development packs, including CPU-reference exports, and AXDocLayout FP32 CUDA TF32 sources. Runtime evidence is model-specific.
-
AutomatosX/AXDocLayout-M-v1.3-CUDA-TF32
Object Detection • 36.7M • Updated • 36 -
AutomatosX/AXDocLayout-M-v1.2-CUDA-TF32
Object Detection • 36.7M • Updated • 134 -
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 35B • Updated • 36 -
AutomatosX/AX-Qwen3.8-27B-CUDA-AXQ-NVFP4-W4A16-MTP
Image-Text-to-Text • 28B • Updated • 106
Low-precision reference storage, including source-layout MX matrices. See reconstruction and runtime limits in each card; not quality-certified.
Muse Glimmer NVFP4 W4A16 development conversion. Vision stays at source precision. Not quality-certified.
AX BI selected AXQ Qwen and Nemotron MLX MXFP4/MXFP8 packs with protected MTP. Development artifacts; see individual model cards.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93
AX Work selected AXQ Qwen and Nemotron MLX MXFP4/MXFP8 packs with protected MTP. Development artifacts; see individual model cards.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93
Qwen3.8, Qwen3.6, Qwen3.5, VL, embeddings, and image reference storage.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93
AXQ embedding packs: MLX affine6/MXFP4/MXFP8 and CUDA NVFP4. See task-specific loader, pooling and runtime evidence. Not certified.
-
AutomatosX/AX-Qwen3-Embedding-0.6B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 0.6B • Updated • 44 -
AutomatosX/AX-Qwen3-Embedding-4B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 4B • Updated • 44 -
AutomatosX/AX-Qwen3-Embedding-8B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 8B • Updated • 45 -
AutomatosX/AX-Nemotron-3-Embed-1B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 1B • Updated • 38
Nemotron-3-Embed 1B/8B retrieval and Super-120B/Lightning-30B MTP AXQ packs: MLX MXFP4/MXFP8 and CUDA NVFP4. Not certified.
-
AutomatosX/AX-Nemotron-3-Embed-1B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 1B • Updated • 38 -
AutomatosX/AX-Nemotron-3-Embed-8B-CUDA-AXQ-NVFP4-W4A4
Sentence Similarity • 8B • Updated • 40 -
AutomatosX/AX-Nemotron-3-Super-120B-A12B-CUDA-AXQ-NVFP4-MTP
Text Generation • 124B • Updated • 326 -
AutomatosX/AX-Nemotron-3.5-Lightning-30B-A3B-CUDA-AXQ-NVFP4-MTP
Text Generation • 33B • Updated • 384
Qwen3.8: MLX affine6/MXFP4/MXFP8 and CUDA NVFP4 development packs with packaged MTP.
-
AutomatosX/AX-Qwen3.8-27B-CUDA-AXQ-NVFP4-W4A16-MTP
Image-Text-to-Text • 28B • Updated • 106 -
AutomatosX/AX-Qwen3.8-27B-MLX-AXQ-MXFP4-MTP
Text Generation • 27B • Updated • 853 -
AutomatosX/AX-Swift-1.5-Qwen3.8-27B-MLX-AXQ-MXFP4-MTP
Text Generation • 27B • Updated • 577 • 1 -
AutomatosX/AX-Qwen3.8-Flash-Next-MLX-AXQ-MXFP4-MTP
Text Generation • 177B • Updated • 599
Tiel Coder 35B-A3B affine6/MXFP4/MXFP8 with packaged MTP, and the Cyber variant.
-
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 751 • 1 -
AutomatosX/AX-Cyber-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 797 -
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 236 -
AutomatosX/AX-Cyber-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 323
Coding-source AXQ packs including Tiel, Cyber, Swift, Flash Next and Qwen3-Coder-Next. Check per-model runtime and MTP evidence.
-
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 751 • 1 -
AutomatosX/AX-Tiel-Coder-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 236 -
AutomatosX/AX-Swift-1.5-Qwen3.8-27B-MLX-AXQ-MXFP4-MTP
Text Generation • 27B • Updated • 577 • 1 -
AutomatosX/AX-Swift-1.5-Qwen3.8-27B-MLX-AXQ-MXFP8-MTP
Text Generation • 27B • Updated • 329
Official 35B-A3B MiniCPM-V 4.7 AXQ MLX MXFP4/MXFP8 and NVFP4 development packs. CPU-reference NVFP4; no certification.
-
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-MLX-AXQ-MXFP4
Image-Text-to-Text • 35B • Updated • 121 -
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-MLX-AXQ-MXFP8
Image-Text-to-Text • 35B • Updated • 158 -
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 35B • Updated • 36 -
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-MLX-AXQ-6bit
Image-Text-to-Text • 35B • Updated • 111
Qwen3-VL MLX affine6/MXFP4/MXFP8 and CUDA NVFP4 packs. Protected vision; model-specific image smoke evidence. Not certified.
-
AutomatosX/AX-Qwen3-VL-4B-Instruct-CUDA-AXQ-NVFP4-W4A4
Image-Text-to-Text • 4B • Updated • 38 -
AutomatosX/AX-Qwen3-VL-8B-Instruct-CUDA-AXQ-NVFP4-W4A4
Image-Text-to-Text • 9B • Updated • 37 -
AutomatosX/AX-Qwen3-VL-4B-Instruct-MLX-AXQ-MXFP4
Image-Text-to-Text • 4B • Updated • 70 -
AutomatosX/AX-Qwen3-VL-4B-Instruct-MLX-AXQ-MXFP8
Image-Text-to-Text • 4B • Updated • 67
Cloudflare Clef, Clef-Flash, and Clef-Omni: MLX MXFP4/MXFP8 and CUDA NVFP4 W4A16.
-
AutomatosX/AX-Clef-Flash-9B-CUDA-AXQ-NVFP4-W4A16
Text Generation • 9B • Updated • 36 -
AutomatosX/AX-Clef-27B-CUDA-AXQ-NVFP4-W4A16
Text Generation • 27B • Updated • 43 -
AutomatosX/AX-Clef-Omni-30B-A3B-CUDA-AXQ-NVFP4-W4A16
Text Generation • 35B • Updated • 38 -
AutomatosX/AX-Clef-Flash-9B-MLX-AXQ-MXFP4
Text Generation • 9B • Updated • 54
DeepSeek-OCR-2 MLX MXFP8 and CUDA NVFP4 W4A16/W4A4 development packs.
-
AutomatosX/AX-DeepSeek-OCR-2-CUDA-AXQ-NVFP4-W4A4
Image-Text-to-Text • 3B • Updated • 51 -
AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-MXFP8
Image-Text-to-Text • 3B • Updated • 144 -
AutomatosX/AX-DeepSeek-OCR-2-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 3B • Updated • 54 -
AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-6bit
Image-Text-to-Text • 3B • Updated • 99
DeepSeek-OCR-2 and Unlimited-OCR v1/v2 development packs. Runtime evidence is pack-specific; see each model card.
-
AutomatosX/AX-DeepSeek-OCR-2-CUDA-AXQ-NVFP4-W4A4
Image-Text-to-Text • 3B • Updated • 51 -
AutomatosX/AX-Unlimited-OCR-3B-MoE-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 3B • Updated • 51 -
AutomatosX/AX-DeepSeek-OCR-2-MLX-AXQ-MXFP8
Image-Text-to-Text • 3B • Updated • 144 -
AutomatosX/AX-DeepSeek-OCR-2-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 3B • Updated • 54
Unlimited-OCR v1 development packs and v2 MXFP4/MXFP8/NVFP4 reference storage. Runtime evidence is pack-specific.
-
AutomatosX/AX-Unlimited-OCR-3B-MoE-CUDA-AXQ-NVFP4-W4A16
Image-Text-to-Text • 3B • Updated • 51 -
AutomatosX/AX-Unlimited-OCR-3B-MoE-MLX-AXQ-MXFP8
Image-Text-to-Text • 3B • Updated • 264 -
AutomatosX/AX-Unlimited-OCR-3B-MoE-MLX-AXQ-6bit
Image-Text-to-Text • 3B • Updated • 138 -
AutomatosX/AX-Unlimited-OCR-2-CUDA-AXQ-NVFP4-W4A16
3B • Updated • 55 • 1
Hy-MT2 MLX MXFP4/MXFP8 and CUDA NVFP4 W4A16. Not certified.
-
AutomatosX/AX-Hy-MT2-30B-A3B-CUDA-AXQ-NVFP4-W4A16
Translation • 30B • Updated • 66 -
AutomatosX/AX-Hy-MT2-30B-A3B-MLX-AXQ-MXFP4
Text Generation • 30B • Updated • 277 -
AutomatosX/AX-Hy-MT2-30B-A3B-MLX-AXQ-MXFP8
Text Generation • 30B • Updated • 99 -
AutomatosX/AX-MiniCPM-V-4.7-35B-A3B-MLX-AXQ-MXFP4
Image-Text-to-Text • 35B • Updated • 121
MLX packs only, including model-specific detectors. Native NVIDIA checkpoints are listed in the CUDA collection.
-
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP4-MTP
Text Generation • 35B • Updated • 189 -
AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-MXFP8-MTP
Text Generation • 35B • Updated • 75 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP4-MTP
Text Generation • 9B • Updated • 187 -
AutomatosX/AX-Qwen3.5-9B-MLX-AXQ-MXFP8-MTP
Text Generation • 9B • Updated • 93