Views
No views yet
Licensing note: the bundled models are not all under a single license. Four are Apache-2.0 and two (Silero VAD, BEATs) are MIT. See THIRD_PARTY_LICENSES.md for the per-model attribution and verbatim license texts.
.tar (tracked with Git LFS) containing a single
top-level *.mlpackage directory — an unflattened Core ML model package
with its Manifest.json, the model.mlmodel spec, and the
weights/weight.bin blob. Unpack before use:tar xf yolox_l_1024x640_w8.tar # -> yolox_l_1024x640_w8.mlpackage/.mlpackage via its model_path
configuration. In the filenames, w8 denotes 8-bit weight quantization
(activations run in FP16 on the Apple Neural Engine); resolutions are
written WxH.| Archive | Model | Task | License |
|---|---|---|---|
qwen3_5_mlx_4b_vision_vid_512x320_w8.tar | Qwen3.5-4B vision tower @ 512×320 (Core ML) | Vision encoder for a multimodal LLM | Apache-2.0 |
qwen3_5_mlx_4b_vision_vid_768x480_w8.tar | Qwen3.5-4B vision tower @ 768×480 (Core ML) | Vision encoder for a multimodal LLM | Apache-2.0 |
gemma4_mlx_e4b_vision_768x480_w8.tar | Gemma 4 (e4b) vision tower @ 768×480 (Core ML) | Vision encoder for a multimodal LLM | Apache-2.0 |
yolox_l_1024x640_w8.tar | YOLOX-L (Core ML) | Object detection | Apache-2.0 |
silero-vad-unified-v6.tar | Silero VAD v6, unified (Core ML) | Voice activity detection | MIT |
beats_as2m_10s.tar | BEATs iter3+ (AS2M, cpt1) (Core ML) | Audio tagging / sound-event classification | MIT |
qwen3_5_mlx_4b_vision_vid_512x320_w8variant=tensor_video_rgb), the vision encoder
that turns a pair of video frames into the embedding sequence consumed by
the Qwen3.5-4B language model.image0, image1 — the two consecutive video frames that
form one temporal patch, each a planar RGB fp16 tensor (1, 3, 320, 512)
(1×3×H×W, i.e. 512×320 W×H) with raw pixels in 0..255. The
x/127.5 - 1 normalisation is baked into the graph (applied on the ANE).image_features — vision features of shape (160, 2560), fp16.torch==2.12.0).qwen3_5_mlx_4b_vision_vid_768x480_w8variant=tensor_video_rgb) at a higher
768×480 input — identical I/O contract to the 512×320 build, with a longer
output token sequence.image0, image1 — the two consecutive video frames that
form one temporal patch, each a planar RGB fp16 tensor (1, 3, 480, 768)
(1×3×H×W, i.e. 768×480 W×H) with raw pixels in 0..255. The
x/127.5 - 1 normalisation is baked into the graph (applied on the ANE).image_features — vision features of shape (360, 2560), fp16.torch==2.12.0).gemma4_mlx_e4b_vision_768x480_w8image — a single planar RGB fp16 MLMultiArray
[1, 3, 480, 768] (1×3×H×W, i.e. 768×480 W×H), raw pixels in 0..255
(channel order R, G, B). Preprocessing is baked in: x/255 then
2*(x - 0.5).soft_tokens — shape [1, 160, 2560], fp16.torch==2.12.1).yolox_l_1024x640_w8image — Core ML image input at 1024×640.output — detection tensor [1, N, 4 + num_classes]
((cx, cy, w, h) in model-input pixels plus sigmoid class scores).torch==2.12.1); all
operations placed on the Apple Neural Engine (no CPU fallback ops).silero-vad-unified-v6audio_input (mono PCM window) plus the recurrent
hidden_state / cell_state carried between calls.vad_output (speech probability) plus
new_hidden_state / new_cell_state.torch==2.7.0).beats_as2m_10scpt1 checkpoint), converted to Core ML over a fixed 10-second window. It
classifies audio into the AudioSet sound-event ontology.waveform — a mono 16 kHz waveform over a fixed 10-second
window (160,000 samples).probs — per-class probabilities over the AudioSet ontology
(527 sound-event classes).microsoft/unilm) — MIT.torch==2.7.0).ed27924a97a873a019d1c8a1e1f9826bd30da2588683843d6c59cfb0b2289f30 qwen3_5_mlx_4b_vision_vid_512x320_w8.tar
3e928ea0c3e6848896278b68436d46f1aa306d95d8ca4a1faadfad354d705b87 qwen3_5_mlx_4b_vision_vid_768x480_w8.tar
17bebe9fdb416a760df4d00ed39c7edf296ba768d54139c8183c1323da646d3e gemma4_mlx_e4b_vision_768x480_w8.tar
7f7bf85947bb228f978470284e16a9d00c906a6cb07359f9a51c68806fa9c609 yolox_l_1024x640_w8.tar
ee14d006d1cea87f15a4f0ac5f3e1e00b80a10c40eca855e8ff6741f389b4423 silero-vad-unified-v6.tar
47c63f83bfa47e248b199e820ae6a01a5ed40a5f6ef4b85f42f12d7ac5754638 beats_as2m_10s.tar