Downloads · 30 days
104
5% of all-time downloads
alessandrobologna/DeepSeek-V4-Pro-DSpark-Drafter-GGUF
DeepSeek-V4-Pro-DSpark-Drafter-GGUF is a machine learning model from alessandrobologna. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for gguf. The card lists the license as mit.
This repository contains only the DSpark auxiliary drafter for DeepSeek V4 Pro. It is not a standalone language model and does not contain the 893 GB target checkpoint.
Downloads · 30 days
104
5% of all-time downloads
All-time downloads
2.3K
Public
Repo size
79.9 GB
Likes
2
Public
Click a slice to open those files.
.gguf53.2 GB · 100%
From the Hugging Face model README
This repository contains only the DSpark auxiliary drafter for DeepSeek V4 Pro. It is not a standalone language model and does not contain the 893 GB target checkpoint.
The GGUF is built from the official
deepseek-ai/DeepSeek-V4-Pro-DSpark
checkpoint using the standalone Python scripts included here. The source
manifest, deterministic tensor recipe, structural parser, numeric verifier,
dependency pin, and build evidence are published alongside the artifact.
| File | Bytes | SHA-256 |
|---|---|---|
DeepSeek-V4-Pro-DSpark-Drafter-Q2_K-Q8_0.gguf | 26,616,105,472 | 1ddd0ad10583658809ec6b8c69b02214c83e1d2e0cd89e0c197d86eda10cede7 |
DeepSeek-V4-Pro-DSpark-Drafter-Q2_K-Q8_0-dflash.gguf | 26,621,353,856 | 9e53ffd4fc6bd298ad4d1428f7da1d88615b266294ef2f1f4f95b9ac1279169a |
The legacy artifact uses GGUF v3 and the custom
deepseek_v4_pro_dspark_draft architecture. Its 81 tensors use this policy:
| Output type | Tensors | Transformation |
|---|---|---|
| F32 | 45 | BF16 expanded exactly to F32, or source F32 bytes preserved |
| F16 | 2 | BF16 Markov heads rounded to F16 |
| Q8_0 | 25 | FP8 E4M3 weights with E8M0 block scales dequantized, then quantized to Q8_0 |
| Q2_K | 9 | Routed-expert MXFP4 decoded and requantized with the unweighted DS4 reference Q2_K algorithm |
The F16, Q8_0, and Q2_K transformations are lossy and are checked against explicit numeric error bounds. This repository does not claim that the complete conversion is lossless.
deepseek-ai/DeepSeek-V4-Pro-DSpark7c09739fd136abfb70a49ec334157f65f45b52cdThe three tail shards contain the complete DSpark module and no target-model
tensors. The other 63 shards are not downloaded. Exact input sizes and
SHA-256 values are in manifest/source.json; the
downloader and converter reject any mismatch.
The reference build uses Linux aarch64, Python 3.14.6, uv 0.11.29, and NumPy
2.5.1 from the hash-pinned binary wheel in
requirements-linux-aarch64-py314.lock. Source download requires the hf
CLI; the reference build used version 1.24.0. The upstream repository is
public, so a Hugging Face token is not required.
uv venv --python 3.14.6 .venv
uv pip install --python .venv/bin/python \
--require-hashes --only-binary=:all: \
-r requirements-linux-aarch64-py314.lock
.venv/bin/python scripts/download_sources.py --destination sources
.venv/bin/python -m unittest discover -s tests -v
.venv/bin/python scripts/reproduce.py \
--sources sources \
--recipe q2_k-q8_0 \
--manifest-dir manifest \
--repeat-check
sha256sum -c SHA256SUMS
Each downloaded byte is checked against manifest/source.json before use.
--repeat-check performs a second clean conversion and requires the complete
GGUF SHA-256 to match byte for byte.
The converter memory-maps the source shards, processes FP8 tensors in 128-row bands, and uses at most sixteen ordered worker threads for Q2_K. Each worker keeps at most one decoded routed expert resident, and results are written in expert order so scheduling cannot affect the GGUF bytes. The build does not require PyTorch, CUDA, safetensors, a GGUF package, or a GPU.
The included verifier checks:
Machine-readable evidence is written to:
manifest/build-q2_k-q8_0.jsonmanifest/validation-q2_k-q8_0.jsonSHA256SUMSThe Q2_K verifier enforces a maximum block-relative reconstruction error of
0.55; the measured maximum was 0.5107014775276184. Q8_0 enforces 0.01;
its measured maximum was 0.00396728515625. The Pro Q2_K bound covers the
measured maximum that exceeded the Flash recipe's 0.50 bound; it is not a
claim about end-to-end model quality. The focused unit test also requires
byte-for-byte agreement with a nontrivial 84-byte block emitted by the
attributed C reference quantizer.
DeepSeek-V4-Pro-DSpark-Drafter-Q2_K-Q8_0-dflash.gguf uses GGUF v3,
general.architecture=dflash, canonical llama.cpp metadata and tensor names,
and canonical one-based dflash.target_layers=[59, 60, 61].
The schema is pinned to ggml-org/llama.cpp commit
3581ba0cf591b3f772fbb002de0f70e294bc0396, incorporating PR #25784 and
the separate-DSpark converter in PR #26452. Its exact type inventory is F32
45, F16 2, Q8_0 25, Q2_K 9, and BF16 0.
The standardized file preserves every tensor payload from the custom
Q2_K/Q8_0 artifact byte for byte through an explicit checked name mapping. It
is not a new quantization. Its complete tokenizer metadata was copied from the
compatible target through the official llama.cpp DeepSeek V4 DSpark converter
at commit 15586e2d7165570fb3aa7c26e0d442e289ef69de, including
tokenizer.ggml.mask_token_id=128799. The legacy artifact remains available.
Reproduce and verify the standardized candidate with:
.venv/bin/python -m unittest discover -s tests -v
.venv/bin/python scripts/dflash.py \
--sources sources \
--target-tokenizer-gguf tokenizer.gguf \
--manifest-dir manifest \
--repeat-check \
--legacy-reference DeepSeek-V4-Pro-DSpark-Drafter-Q2_K-Q8_0.gguf
sha256sum -c SHA256SUMS
Dedicated build, validation, payload-comparison, and pinned-schema manifests
are published under manifest/.
deepseek_v4_pro_dspark_draft and
dspark.* dialect. The standardized file requires llama.cpp dflash
support.ns4 standardized dflash path supports Flash only. ns4
needs Pro runtime support before it can load this standardized Pro file.DeepSeek-V4-Pro-IQ2XXS-w2Q2K-AProjQ8-SExpQ8-OutQ8-Instruct-imatrix.gguf
at revision a88c423b511666d7ff7a4dcaee651669312bea97, SHA-256
a0314d9c0e16122cd60071079124a2d17185d317c55a8f95ecb3ed3506278a96.The pinned source defines:
These values are embedded in the GGUF metadata and validated during the build.
The source-derived weights remain under DeepSeek's MIT terms in
LICENSE. The conversion and verification scripts are under
LICENSE.code. The pinned checkpoint's
upstream license
and the DS4/GGML attribution in
THIRD_PARTY_NOTICES.md provide the applicable
notices.