Minimal inference

A readable reference implementation rather than a production serving engine. The model code covers the vision encoder and aligner, sliding-window plus compressed sparse attention with its two-level indexer, engram n-gram lookups, MoE, Hyper-Connections, and the DSpark forward path. Generation itself is plain autoregressive sampling.

Install

python -m pip install -r requirements.txt

Convert Hugging Face weights

The runtime uses one converted checkpoint file per tensor-parallel rank. From this directory:

export HF_CKPT_PATH=/path/to/DeepSeek-V4.1-Flash-HF
export SAVE_PATH=/path/to/DeepSeek-V4.1-Flash-TP8
export MP=8

python convert.py \
  --hf-ckpt-path "${HF_CKPT_PATH}" \
  --save-path "${SAVE_PATH}" \
  --model-parallel "${MP}" \
  --expert-dtype fp4 \
  --tokenizer-path "${HF_CKPT_PATH}"

Expert counts are inferred from the weight names, so they do not need to be passed. --tokenizer-path points at whichever directory holds tokenizer.json and tokenizer_config.json; they are copied into the converted checkpoint.

Run the equivalent TXT and JSON examples

export CKPT_PATH=/path/to/DeepSeek-V4.1-Flash-TP8
export MP=8

INPUT_FILE=examples/example.txt ./run.sh
INPUT_FILE=examples/example_harmony.json ./run.sh

The two files express the same interleaved two-image prompt, so they produce identical encoded prompts and input token IDs.

For interactive chat:

torchrun --nproc-per-node "${MP}" generate.py \
  --ckpt-path "${CKPT_PATH}" \
  --config config.json \
  --interactive \
  --temperature 0.6

For multi-node execution, pass the usual torchrun --nnodes, --node-rank, --master-addr, and --master-port arguments before generate.py.

Self-test

model.py builds a small model from the ModelArgs defaults and runs a prefill plus 22 decode steps, exercising the real dense-fp8 / MoE-fp4 kernels. Weights are uninitialized, so it checks shapes and kernel plumbing, not numerics:

python model.py