MLX files for ZDTaichu5.0-9B, an advanced multimodal foundation model developed by the Zi Dong Tai Chu team for visual understanding, spatial reasoning, agentic tool use, and embodied AI workloads. Built on a Qwen3.5-9B language backbone and C-RADIOv4-H vision encoder, the source checkpoint supports a native context length of 131,072 tokens (128K).
Benchmark results reported by TaichuAI for the original ZDTaichu5.0-9B foundation model. These figures are not measurements of this MLX conversion.
| Quantization | Repository | Size (GB) |
|---|---|---|
| 8-bit | ZDTaichu5.0-9B-MLX-8bit | 11.20 GB |
| 6-bit | ZDTaichu5.0-9B-MLX-6bit | 8.96 GB |
| 4-bit | ZDTaichu5.0-9B-MLX-4bit | 6.72 GB |
| Component | Architecture | Precision |
|---|---|---|
| Language Backbone | Qwen3.5 hybrid (Gated DeltaNet + full attention, 3:1 ratio) | Quantized (6-bit affine, group_size=64) |
| Vision Encoder | C-RADIOv4-H (ViT-H/16, 653M) | BF16 (full precision) |
| Projector | RMSNorm → Linear(5120→20480) → SquaredReLU → Linear(20480→4096) | BF16 (full precision) |
The vision encoder and multimodal projector are kept at full precision (BF16) to prevent visual reasoning degradation.
The MLX conversion embeds the upstream chat template. An external copy is provided as chat_template.jinja for runtimes that require a separate template file.
mlx-vlmmlx_vlm.generate \
--model abenzerps/ZDTaichu5.0-9B-MLX-6bit \
--image path/to/image.jpg \
--prompt "Describe what is shown in this image in detail." \
--max-tokens 512 \
--temp 0.7
mlx-vlmmlx_vlm.generate \
--model abenzerps/ZDTaichu5.0-9B-MLX-6bit \
--prompt "Explain why reproducible builds matter." \
--max-tokens 512 \
--temp 0.7
from mlx_vlm import load, generate
model_id = "abenzerps/ZDTaichu5.0-9B-MLX-6bit"
model, processor = load(model_id)
prompt = "Describe what is shown in this image in detail."
image = "path/to/image.jpg"
output = generate(model, processor, prompt=prompt, image=image, max_tokens=512)
print(output)
Increase context up to 131,072 tokens (128K) when sufficient unified memory is available on Apple Silicon. Tool-call behavior depends on the serving runtime and its parser integration; use the embedded template and verify tool calls in the target application.
model-00005-of-00005.safetensors 2.5 GBmodel-00001-of-00005.safetensors 1.7 GBmodel-00002-of-00005.safetensors 1.7 GBmodel-00003-of-00005.safetensors 1.7 GBmodel-00004-of-00005.safetensors 1.3 GBHave the files? Submit a torrent.
Submit a torrentHave the complete files and permission to share them? Replace the example folder path. The script checks the files, makes a torrent and submits its magnet and checksums. Keep your torrent client seeding. Get your community key from your settings, and read the script before running it.
curl -fsSL https://pirateface.co/package.sh | bash -s -- --repo abenzerps/ZDTaichu5.0-9B-MLX-6bit ./model-folderTaichuAI/ZDTaichu5.0-9B