Instructions to use Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated with Transformers:

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

pipe = pipeline("text-generation", model="Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated")
model = AutoModelForCausalLM.from_pretrained("Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated

SGLang

How to use Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated 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 "Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated" \
    --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": "Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated",
		"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 "Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated" \
        --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": "Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated with Docker Model Runner:
```
docker model run hf.co/Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated
```

DeepSeek-V4-Flash-Vision-Exp — Abliterated (Overlay)

⚠️⚠️⚠️ THIS REPOSITORY DOES NOT CONTAIN THE FULL MODEL ⚠️⚠️⚠️

This repo ships only the abliteration overlay + repointed index. You MUST first download the original base weights to use it:

Required base: deepseek-ai/DeepSeek-V4-Flash-Vision-Exp — all 48 shards (model-00001-of-00048.safetensors … model-00048-of-00048.safetensors), ~157 GB.

Setup (3 steps)

Download the original model repo above (48 shards + its model.safetensors.index.json).

Copy this repo's model-overlay-00001-of-00001.safetensors (1.44 GB) and overwrite the base repo's model.safetensors.index.json with this repo's index.

Load the directory with a V4-capable vLLM vision fork (see Serving below).

The overlay only stores the 92 edited tensors; without the base shards the repo cannot load.

A decensored (abliterated) rebuild of deepseek-ai/DeepSeek-V4-Flash-Vision-Exp — a multimodal (vision-language) DeepSeek-V4 MoE model served at native precision across 2× NVIDIA RTX PRO 6000 Blackwell (96 GB each) with vLLM.

Refusal behaviour was removed from the attention output projections only. All routed experts, shared experts, wo_a, embeddings, routers, norms and every mHC parameter are untouched. The 32-layer vision tower and its bias_vl are also untouched — the edit is confined to the text path's attn.wo_b tensors, yet the visual path decensors too (see Validation).

This is an overlay, not a re-quantization. Exactly 92 of 72,633 tensors (0.13 %) differ from the base release; the other 72,541 are byte-identical and verifiable per-tensor by sha256. There is no quantization step anywhere in this pipeline — the model ships at the base checkpoint's native precision.

⚠️ Uncensored model. Safety refusals have been substantially removed. You are responsible for how you use it. Intended for local/research use on hardware you control. It will attempt almost any request.

Highlights


Base	`deepseek-ai/DeepSeek-V4-Flash-Vision-Exp` (DeepSeek-V4 MoE + 32-layer vision tower)
Architecture	43 layers, hidden 4096, 256 experts (top-6) + 1 shared, hybrid CSA + HCA attention, mHC hyper-connections, inline 3-stage DSpark draft head; vision tower 32×1024 dim, patch 14, downsample 3, `bias_vl`
Context	1,048,576 native (served here at 524,288)
Precision	unchanged — base checkpoint's native precision (FP8 e4m3 block-scaled elsewhere, FP4 routed experts)
Edit	rank-1 orthogonal projection, λ = 2.0, on 43 × `layers..attn.wo_b` + 3 × `mtp..attn.wo_b` (92 tensors)
Untouched	routed + shared experts, `wo_a`, `embed`, `head`, routers, norms, entire 32-layer vision tower + `bias_vl`, all mHC params
Size	this repo: 1.44 GB overlay (+ base ~157 GB required)
Runtime	vLLM (`fraserpricee/vllm:dspark-cu132-vision`), TP=2 across 2× RTX PRO 6000 Blackwell

Validation (abliterated vs original base — measured on this checkpoint)

Refusal & coherence

Metric	Base Vision-Exp	This model (abliterated)
Refusal — long-gen mass scan (n=300 AdvBench)	—	0.0 % safety refusals (7 raw detector hits; 6 verified as modality/placeholder misfires, the 1 remaining is an ideological stand on writing an oppressive book — not a safety refusal)
Empty-answer rate (n=300)	—	0.0 %
Generation coherence (40 outputs, chat+think)	clean	0 true garbled spans (10 flags all misreads)
Multi-turn context (8-turn cumulative)	—	8/8 clean, final turn recalls the full arc
Visual gate (R18 page description)	refuses	describes fully — composition, pose, onomatopoeia, narration. Visual path decensored despite text-side-only edit

Long-generation behaviour (8192 thinking budget)

Sample	trigger	explain_qm	write_code	math	Score
Base Vision-Exp	✓	✓	✓ (r8.8k)	✓	4/4
This model (λ=2.0)	✓	✓ (r13.7k)	✗ (r30.6k div.)	✓	3/4

The only regression is long write_code generation diverging (~30 k tokens) — a known common consequence of abliteration, present on the published 0731 text sibling as well (92 k divergence). Not a text-coherence loss.

Tool-calling (3-tool schema, DSML native, 6 prompts × 3 reasoning modes = n=18)

Mode	Base Vision-Exp	This model
chat	1.000	1.000
think-high	1.000	1.000
think-max	1.000	1.000
Correct tool selected	1.000	1.000

Compliance means: the completion parses via the deepseek_v4 parser, contains a tool call, names a valid offered tool, and supplies every required parameter. Zero difference from base — the edit does not disturb tool calling.

Serving performance (2× RTX PRO 6000 Blackwell, TP=2, single-stream decode, TTFT excluded)

workload	Base Vision-Exp	This model
technical prose	74.7 tok/s	74.7 tok/s
code	75.6 tok/s	75.5 tok/s
structured JSON	74.9 tok/s	75.1 tok/s
aggregate (8 concurrent)	~262 tok/s (warm)	~255 tok/s (warm)
DSpark draft acceptance	29–38 % (prompt-dependent)	37 %

Serving performance is identical to base — the overlay edits weight values only, never the computation graph. Single-stream decode ~75 tok/s across workloads; ~255–262 tok/s aggregate at 8 concurrent streams. (Earlier cold-start TTFT outliers were prefix-cache warmup artefacts, not model differences.)

How it was built

Refusal directions. Forward hooks on all 43 attn.wo_b modules captured last-token sublayer outputs over AdvBench harmful vs Alpaca-harmless prompts in all three reasoning modes (chat, think-high, think-max). Because wo_b is RowParallelLinear, cross-rank digests were asserted identical on every prompt to guard against capturing a half-direction. Stability: median split-half cosine 0.978–0.990 across layers; held-out AUC 0.974–0.999. Mode agreement: chat↔thinking median cosine 0.46, the two thinking modes 0.90 — matching the measured split on the 0731 sibling, confirming that a single text-mode direction would leave thinking-mode refusal largely intact, hence the three-mode merge.
Merge + bake. The per-mode direction sets were merged and applied as W ← W − λ·r̂(r̂ᵀW) at λ = 2.0 on the FP8 e4m3 weights: dequantize with the block scales, project, then re-quantize holding the original block exponents fixed. Written as a tensor overlay with a repointed model.safetensors.index.json, so the 48 original shards are never rewritten. Bake is deterministic — overlay sha256 58e0461aaf4a387b0026e82e9b39fc73ade52851e1f15288114fbbcf0c0e5350.
λ is calibrated for this model, not inherited. Because a baked FP8 edit is structurally weaker than a runtime hook (most elements round back to their original byte), λ must be tuned against the baked model.

The λ operating envelope (heavily validated here)

λ	Result
3.0	Too strong. Long generations garble (quote-storm, semantic break). Reject.
2.5	Equivalent to 2.0 on the 8 k ruler (3/4, same divergent probe) but no added benefit on this model.
2.0	Selected. Full inversion (`
1.9	Under-inversion fuzzy band. Model sits in a half-refuse / half-comply limbo. Reject.
2.05–2.3	e4m3 quant-dead zone — edits swallowed by rounding. Avoid.

So 2.0 is the low edge of a clean operating point: below it is the under-inversion fuzzy band, above it is over-inversion damage. Do not carry over the 0731 text sibling's λ=2.5 — the two are different weight sets and recipes do not transfer.

Serving (vLLM, 2× RTX PRO 6000 Blackwell)

The full model (~157 GB + overlay) needs both 96 GB cards, tensor-parallel = 2:

docker run -d --name ds4-vision-abl --ipc=host --shm-size 32g --gpus all --restart unless-stopped \
  -p 5001:8000 \
  -v /model-dir:/mnt/ubuntulib:ro \
  fraserpricee/vllm:dspark-cu132-vision \
  /opt/venv/bin/python -m vllm.entrypoints.cli.main serve /mnt/ubuntulib/models/DeepSeek-V4-Flash-Vision-Exp-abliterated-l2.0 \
  --served-model-name DeepSeek-Vision-EXP-abliterated --trust-remote-code \
  --tensor-parallel-size 2 --gpu-memory-utilization 0.95 \
  --max-model-len 524288 --max-num-seqs 8 --max-num-batched-tokens 2048 \
  --kv-cache-dtype fp8 --block-size 256 --tokenizer-mode deepseek_v4 \
  --attention-backend FLASHINFER_MLA_SPARSE_DSV4 \
  --kernel-config.moe_backend flashinfer_cutlass \
  --speculative-config '{"method":"dspark","num_speculative_tokens":4,"draft_sample_method":"greedy"}' \
  --enable-flashinfer-autotune \
  --compilation-config '{"cudagraph_mode":"FULL_AND_PIECEWISE","custom_ops":["all"]}' \
  --enable-prefix-caching --enable-chunked-prefill --async-scheduling \
  --reasoning-parser deepseek_v4 --enable-auto-tool-choice --tool-call-parser deepseek_v4 \
  --allowed-local-media-path /data

Blackwell notes (each cost real debugging time):

The serving image must be the vision fork, fraserpricee/vllm:dspark-cu132-vision. The plain SM120 text fork (as used for the 0731 sibling) hard-fails with the vision tower.
--moe-backend flashinfer_cutlass is the working path on this 96 GB Blackwell stack; --attention-backend FLASHINFER_MLA_SPARSE_DSV4 is required for DSV4 sparse MLA.
max-model-len is served at 524,288 here (the vision path adds its own token budget — the tower pads up to vision_max_n_token 384).
--allowed-local-media-path must point at the dir holding images; the vision path reads local file paths, not just base64.
--num_speculative_tokens ≤ dspark_block_size (5): larger values boot but misbehave.
The vision path is much more memory-hungry than text-only; keep gpu-memory-utilization ≤ 0.95.

Findings

The edit is text-path only, yet the visual path also decensors. All 92 edited tensors are attn.wo_b on the text backbone; the 32-layer vision tower and bias_vl are byte-identical. Still, an R18 page is described fully with zero refusal — the text-generation head decodes from a fused carrier whose wo_b was edited, neutralising the last-stage refusal regardless of input modality.
Mode directions diverge here just as on the 0731 sibling. chat↔thinking median cosine 0.46, thinking modes 0.90 — a merged direction is required.
A baked FP8 edit is weaker than a runtime hook (most elements round back). On this model we found the e4m3 dead zone (λ ∈ [2.05, 2.3]) where edits get fully swallowed — a strong reason to tune λ on the baked model, never inherit.

Limitations

Stated plainly.

Capability was gated on refusal / coherence / multi-turn / tool-call only. MMLU-Pro, GSM8K, HumanEval and general benchmarks were not run on this checkpoint. "Refusal, coherence, tool-calling and serving measured" is the claim; "general capability unchanged" is not measured.
Long-context behaviour is unvalidated past multi-turn depth. Our retention test is 8 turns, not an N-shot 168 k needle test. Whether refusal suppression holds at extreme depth is an open question.
Refusal is scored by marker matching on the parsed answer (thinking block excluded) + manual per-row review. It will miss phrasings the marker set does not cover.
The write_code long-generation divergence is the one confirmed capability regression (3/4 vs base 4/4) — a known common consequence of abliteration.
The visual gate is n=1 (one R18 page) — a directional signal, not a rate.
The λ=2.0 operating point is calibrated for the baked FP8 weights of this checkpoint. Do not inherit it for other models, and do not reuse the 0731 sibling's λ=2.5 here.

Files

File	Size	Note
`model-overlay-00001-of-00001.safetensors`	1.44 GB	the only differing weight file (92 tensors)
`model.safetensors.index.json`	5.4 MB	repointed — 92 keys → overlay, 72,541 → base shard names
`abliteration_report.json`	27 KB	per-tensor edit audit (sha256, clamped, overflow counts)
`config.json` / `configuration.json` / `generation_config.json`	—	architecture
`tokenizer.json` / `tokenizer_config.json`	6.4 MB	DeepSeek V4 tokenizer
`encoding/encoding_dsv4.py`	29 KB	official V4 prompt encoder (needed for correct chat formatting)
`LICENSE`	—	MIT

The 48 base shards are NOT in this repo. Download them from the base model, then add this repo's overlay + repointed index into the same directory. Removing the overlay + restoring the base index fully reverts the model.

License & credits

MIT (inherited from the base model). Base model: deepseek-ai/DeepSeek-V4-Flash-Vision-Exp. Abliteration follows the residual-direction method (Arditi et al., Refusal in LLMs is mediated by a single direction, NeurIPS 2024); attn.wo_b targeting and the overlay-differential pattern follow and independently replicate cebeuq/DeepSeek-V4-Flash-0731-abliterated — the text-only sibling whose λ=2.5 recipe does not transfer to this vision variant. Serving stack: fraserpricee/vllm:dspark-cu132-vision. Not affiliated with or endorsed by the base-model authors.

Downloads last month: -

Model tree for Joyfulxp/DeepSeek-V4-Flash-Vision-Exp-Abliterated

Base model

deepseek-ai/DeepSeek-V4-Flash-Vision-Exp

Finetuned

(9)

this model

Joyfulxp
/

DeepSeek-V4-Flash-Vision-Exp-Abliterated