---
base_model:
- deepseek-ai/DeepSeek-V4-Flash-0731
- apetersson/DeepSeek-V4-Flash-0731-Abliterated-FP8
pipeline_tag: image-text-to-text
license: mit
tags:
- vision-language
- deepseek-v4
- moonvit
- nvfp4
- fp8
- abliterated
---

# DeepSeek-V4-Flash-0731 Abliterated + Vision (NVFP4)

Abliterated DeepSeek-V4-Flash-0731 (MOE, 284B total / 13B activated) with a
frozen **MoonViT-3d** vision tower (from Kimi-K2.6) and WebBrain's trained
**PatchMerger** projector — packaged for vLLM.

## What this repo is

This is a **fork of `sakamakismile/DeepSeek-V4-Flash-0731-Abliterated-NVFP4`**
(the abliterated NVFP4 conversion) with the vision overlay from
`webbrain-one/DeepSeek-V4-Flash-0731-Vision-NVFP4` spliced on:

| Component | Provenance | Status |
| - | - | - |
| Text checkpoint | sakamakismile ablit 0731 NVFP4 | 48 shards |
| Vision tower | `vision_tower.safetensors` MoonViT | frozen |
| Projector | `mm_projector.safetensors` PatchMerger | frozen |
| Routing | `palette64.json` 64-id palette | see below |

## How to serve with vLLM

The config carries `auto_map` so vLLM's Transformers modeling backend loads it
with remote code — no vLLM fork or custom image tag required:

```bash
docker run --rm --gpus all \
  -p 8000:8000 \
  vllm/vllm-openai:cu129-nightly \
  python -m vllm.entrypoints.openai.api_server \
    --model mumitrol/DeepSeek-V4-Flash-0731-Abliterated-NVFP4-vision \
    --model-impl transformers \
    --trust-remote-code \
    --tensor-parallel-size 4 \
    --dtype bfloat16 \
    --limit-mm-per-prompt image=1
```

- Requires a **Blackwell** node (NVFP4); 1x B300 (288 GB) is enough for a
  single stream / moderate context, 2x for maximum context and throughput.

## Status / validation (IMPORTANT)

This is a **first-pass community port**, not an official NVIDIA or DeepSeek
release. Validate on GPU before production use:

- [ ] Tower attention matches `MMEncoderAttention` (head_dim^-0.5 scale, fused
      wqkv, no RoPE) — F.scaled_dot_product_attention with exact scale.
- [ ] Positional-embedding interpolation mode matches MoonViT (bicubic).
- [ ] Image-to-token seating is right (>= 512-token cap, 2x2 merge layout).
- [ ] Hash-MoE routing for image tokens uses `palette64.json` (deterministic).
- [ ] Abliteration + vision edge effects on refusing behavior are checked.

See `modeling_deepseek_v4_vision.py` for `# ->` implementation notes.

## Licenses

Original DeepSeek weights MIT; Kimi-K2.6-derived tower under its own notice;
WebBrain projector per their repo; abliteration per apetersson/drowzeys notice.
Downstream users must satisfy all applicable licenses.