Instructions to use lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound")

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound

SGLang

How to use lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound with Docker Model Runner:
```
docker model run hf.co/lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound
```

DeepSeek-V4.1-Flash — D1 (Engram tables FP8→MXFP4), auto-round one-command

Same artifact class as DeepSeek-V4.1-Flash-D1-engram-mxfp4 (only the two engram embedding tables requantized FP8→MXFP4), but produced end-to-end by a single auto-round --model_free command (with the DeepSeek-V4.1 source-format patch set).

Checkpoint size: 412.0 GB (vs 510.3 GB official, -19.3%). All 96,081 non-engram tensors verified byte-identical to the official checkpoint; engram tables are int8-packed E2M1 [R,128] + E8M0 [R,8] under their original names, rel_err ≈ 0.12 vs the FP8 source (4-bit E2M1 theoretical band).

Quantization command

AR_MODEL_FREE_SHARD_PARALLELISM=1 \
auto-round --model_name deepseek-ai/DeepSeek-V4.1-Flash \
  --model_free --scheme BF16 \
  --layer_config '{
    "engram.embed":    {"bits": 4, "data_type": "mx_fp4", "group_size": 32},
    "indexer.wq_b":    {"bits": 8, "data_type": "mx_fp8", "group_size": 32},
    "compressor":      {"bits": 16},
    "indexer":         {"bits": 16},
    "shared_experts":  {"bits": 8, "data_type": "mx_fp8", "group_size": 32},
    "ffn.experts":     {"bits": 4, "data_type": "mx_fp4", "group_size": 32},
    "attn":            {"bits": 8, "data_type": "mx_fp8", "group_size": 32},
    "engram.wkv":      {"bits": 8, "data_type": "mx_fp8", "group_size": 32},
    "main_proj":       {"bits": 8, "data_type": "mx_fp8", "group_size": 32}}' \
  --output_dir <out>
# afterwards: restore the official config.json (this repo's config.json is already official)

Notes on the recipe (learned the hard way):

BF16 default keeps every unlisted family in full precision; FP8/MXFP4 source families listed with their same target format pass through byte-identically (no requant).
layer_config patterns are substring matches, first match wins: ffn.experts (not experts) must be used for routed experts so it cannot shadow shared_experts; compressor/indexer are BF16 and must be pinned to bits:16 before attn.
AR_MODEL_FREE_SHARD_PARALLELISM=1 is required for this model: parallel shard workers share one GPU and can OOM the giant-table (384M-row) quantization worker.

Inference (vLLM)

Identical to DeepSeek-V4.1-Flash-D1-engram-mxfp4: vLLM main (≥ PR #56201) + dsv41-quant-plugin engram patch, official config.json (FP8 declaration) kept:

DSV41_ENGRAM_DTYPE=fp4 NCCL_NVLS_ENABLE=0 \
vllm serve <this-model-dir> \
  --tensor-parallel-size 4 --max-model-len 8192 --language-model-only \
  --gpu-memory-utilization 0.90 --port 8100

Provenance

auto-round: patched copy at main @141e4c99 (+270 lines in auto_round/utils/model_free_utils.py: V4.1 source normalize/restore, blockwise FP8 passthrough, chunked MXFP for giant tables, should_skip yields to explicit layer_config).
vLLM main nightly (0.1.1.dev39+g46d2b23ac).
Reference accuracy (same format, script-produced twin): gsm8k 92.19/92.27 vs 92.87 baseline (n=1319, raw 5-shot, thinking off).

Downloads last month: 105

Safetensors

Model size

665B params

Tensor type

BF16

F32

F8_E4M3

Model tree for lvkaokao/DeepSeek-V4.1-Flash-MXFP4-Engram-AutoRound

Base model

deepseek-ai/DeepSeek-V4.1-Flash

Quantized

(64)

this model