deepseek-v4-pro-mtp / README.md
ttupper75035's picture
Add self-describing model card (mtp-sidecar)
75c6b02 verified
|
Raw
History Blame Contribute Delete
1.91 kB
---
base_model: deepseek-ai/DeepSeek-V4-Pro
tags:
- mtp-sidecar
- skulk
- foxlight
license: mit
foxlight:
artifact_type: mtp-sidecar
source_repo: deepseek-ai/DeepSeek-V4-Pro
source_revision: 89d501aed998d33fa4f4702102ec1bb2331e10f6
target_model: deepseek-ai/DeepSeek-V4-Pro
extracted_with: skulk-weights-publisher 0.1.0
generated_at: '2026-06-03T19:21:34Z'
---
# FoxlightAI/deepseek-v4-pro-mtp
This repository contains an **MTP speculative-decoding sidecar** derived from [`deepseek-ai/DeepSeek-V4-Pro`](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro). It is **not a standalone model** — it provides the multi-token-prediction heads used by [Skulk](https://github.com/Foxlight-Foundation/Skulk) to speculatively decode for the target model **`deepseek-ai/DeepSeek-V4-Pro`**. The heads ship at full precision (**bf16, unquantized**) — they are the speculative drafter, where precision drives draft acceptance — so there is **one sidecar per base model** and it serves **every quantization** of the target.
## Provenance
| Field | Value |
| --- | --- |
| Artifact type | mtp-sidecar |
| Source model | `deepseek-ai/DeepSeek-V4-Pro` |
| Source revision | `89d501aed998d33fa4f4702102ec1bb2331e10f6` |
| Target model | `deepseek-ai/DeepSeek-V4-Pro` |
| Extracted with | skulk-weights-publisher `0.1.0` |
| Generated | 2026-06-03T19:21:34Z |
## Usage
Skulk loads this sidecar (`mtp.safetensors`) alongside the target model to enable MTP speculative decoding. It is referenced from the Skulk Weights Publisher catalog and fetched automatically by the Skulk shard downloader; it is not intended to be loaded standalone.
## License
This artifact is derived from [`deepseek-ai/DeepSeek-V4-Pro`](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro) and is published under that model's original license (**mit**), preserved unchanged. Refer to the source model's card for the full terms.