DeepSeek-V4-Pro-Qwen3.5-9B-bf16

mlx-community/DeepSeek-V4-Pro-Qwen3.5-9B-bf16

This model mlx-community/DeepSeek-V4-Pro-Qwen3.5-9B-bf16 was converted to MLX format from Jackrong/DeepSeek-V4-Pro-Qwen3.5-9B using mlx-vlm version 0.4.4.

This is a BF16 MLX conversion. It keeps the source model's chat template and multimodal processor configuration for text/coding, image, and video-style inputs. The model weights were converted to BF16 MLX format; multimodal vision components are preserved.

Refer to the original model card for model details, license, and intended use.

Use with mlx

pip install -U mlx-vlm

Image input

python -m mlx_vlm.generate \
  --model mlx-community/DeepSeek-V4-Pro-Qwen3.5-9B-bf16 \
  --max-tokens 512 \
  --temperature 0.0 \
  --prompt "Describe this image." \
  --image <path_to_image>

Text / coding input

python -m mlx_vlm.generate \
  --model mlx-community/DeepSeek-V4-Pro-Qwen3.5-9B-bf16 \
  --max-tokens 512 \
  --temperature 0.2 \
  --prompt "Write a Python function that parses a JSONL file and counts records by label."

Notes

• This is a BF16 MLX version of Jackrong/DeepSeek-V4-Pro-Qwen3.5-9B.
• The model is intended for Apple Silicon inference with MLX.
• For multimodal usage, prefer mlx-vlm rather than plain mlx-lm.
• License: Apache 2.0, inherited from the source model metadata.

Conversion

mlx_vlm.convert \
  --hf-path Jackrong/DeepSeek-V4-Pro-Qwen3.5-9B \
  --mlx-path DeepSeek-V4-Pro-Qwen3.5-9B-bf16 \
  --dtype bfloat16