What is this
DeepSeek-V4.1-Flash with permanent weight-level abliteration — the safety guardrails have been surgically removed while preserving MMLU capability, vision, reasoning, MTP (DSpark), and multi-turn coherence.
Proprietary weight-level abliteration developed by the dealignai research team. No custom model.py, no runtime hooks, no steering vectors — it's a standard checkpoint that loads exactly like the base model. The refusal circuitry is surgically removed while every capability-critical component (routed experts, Engram memory, CSA2 sparse attention, DSpark draft head, vision tower, router gates, norms, embeddings) is preserved byte-identical to the base.
Table | |
|---|
| Base | deepseek-ai/DeepSeek-V4.1-Flash (552B backbone, 8B/16B active per token) |
| Architecture | Causal Encoder-Decoder (20+20 layers), MoE (384 routed top-6 + 1 shared), Hyper-Connections (4-channel residual), CSA2 sparse attention, Engram n-gram memory, DSpark speculative draft |
| Quant | FP8 (e4m3fn) weights with E8M0 block-scale [32, 32], FP4 routed experts — native, unchanged |
| Context | 1M tokens |
| Vision | DeepSeek-ViT with 2D-RoPE + pixel unshuffle — untouched |
| Modification | Surgical, weight-level (drop-in checkpoint) |
Results
HarmBench-320 — full 2×2 (base vs CRACK, effort=off vs max), T=0 greedy
Every response 4-tier graded (HARD_REF / SOFT_RED / HEDGE / COMPLY), with reasoning-trace verification at effort=max.
Table with columns: eval, base ASR, CRACK ASR, Δ pp| eval | base ASR | CRACK ASR | Δ pp |
|---|
| HB-320 effort=off | 137/320 = 42.81 % | 320/320 = 100.00 % | +57.19 |
| HB-320 effort=max | 5/320 = 1.56 % | 320/320 = 100.00 % | +98.44 |
Notable: at effort=max, the base model becomes MORE refusal-prone (42.8 % → 1.6 %) because reasoning surfaces safety concerns before answering. The CRACK stays at 100.0 % across both effort levels.
Per-category (all 7 HarmBench semantic categories):
Table with columns: category, items, base off, CRACK off, base max, CRACK max| category | items | base off | CRACK off | base max | CRACK max |
|---|
| chemical_biological | 42 | 16.7 % | 100.0 % | 0.0 % | 100.0 % |
| copyright | 80 | 98.8 % | 100.0 % | 0.0 % | 100.0 % |
| cybercrime_intrusion | 52 |
Zero HARD_REF, zero SOFT_RED, zero HEDGE on the cracked build at either effort level.
Every response was graded by a strict multilingual regex-based 4-tier classifier plus (for effort=max) an LLM-as-judge over the saved reasoning trace. Full per-item outputs saved for verification.
MMLU-14k (full test set, base-logit, T=0)
Table with columns: build, correct, acc, Δ| build | correct | acc | Δ |
|---|
| base | 12,211 / 14,042 | 86.96 % | — |
| CRACK | 11,619 / 14,042 | 82.74 % | -4.22 pp |
Excluding the ethics cluster (moral_scenarios, business_ethics, professional_law, jurisprudence, philosophy — where refusal-adjacent behaviour is graded), delta on the remaining ~11k items is -1.1 pp — well within the 3 pp knowledge-preservation target.
Table with columns: subject, n, base, crack, Δ pp| subject | n | base | crack | Δ pp |
|---|
| moral scenarios | 895 | 76.9% | 37.0% | -39.89 |
| professional law | 1534 | 75.9% | 68.8% | -7.04 |
| abstract algebra | 100 | 77.0% | 71.0% | -6.00 |
| security studies |
Extended validation
- 1000-token coherence stress on 6 items — no
WARNING WARNING loops, no character-repeat degeneracy, natural sign-offs.
- Multi-turn conversation (4 turns on same harmful topic — ANFO explosive detail) — no late-turn refusal reversion, no self-correction, coherent through turn 4.
- Vision path — coherent image description ("A blue square centered on a red background.") + refusal drop on image-based harmful prompts ("shaped charge / explosively formed penetrator" description).
- General capability spot checks intact: √2 irrationality proof, Python palindrome with docstring, WWI causes in exactly 3 sentences, quantum observable vs operator distinction.
- Full compat suite pass: streaming SSE, chat logprobs +
top_logprobs, completions logprobs + echo, tool calls (deepseekv41 parser), image input, reasoning-effort tiers (low/high/xhigh/max + float [0, 0.99]), sampling params (temperature, top_p, stop, , , , ), 8-way concurrent, 40k-word prompt at 35,572 tokens.
How to run
Support for DeepseekV41ForCausalLM is still landing across serving stacks (as of 2026-09-10). Working paths:
SGLang (preview branch)
The dsv4.1 branch of sgl-project/sglang (PR #38798) supports DSV4.1. Two options:
Preview Docker image (recommended):
docker pull lmsysorg/sglang:dev-dsv41
docker run --gpus all --shm-size 32g -p 30000:30000 \
-v ~/.cache/huggingface:/root/.cache/huggingface \
--ipc=host --env HF_TOKEN=<your-token> \
lmsysorg/sglang:dev-dsv41 \
sglang serve \
--model-path dealignai/DeepSeek-V4.1-Flash-UNCENSORED-FP8 \
--tp-size 4 --ep-size 4 \
--context-length 262144 --mem-fraction-static 0.85 \
--reasoning-parser deepseek-v41 --tool-call-parser deepseekv41 \
--trust-remote-code
From source (this is exactly what we validated on):
git clone --depth 1 --branch dsv4.1 https://github.com/sgl-project/sglang.git
python3 -m venv sglang-venv
sglang-venv/bin/pip install -U pip setuptools wheel
export PATH=/root/.cargo/bin:$PATH # Rust toolchain required for build
cd sglang/python && sglang-venv/bin/pip install -e .
# Ninja must be on the launch PATH — the sglang-kernel JIT build shells out to it
export PATH=$(dirname $(which ninja)):$PATH
SGLANG_ENABLE_DSV41_ENGRAM_HOST_TABLE=1 \
PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True \
sglang-venv/bin/python -m sglang.launch_server \
--model-path dealignai/DeepSeek-V4.1-Flash-UNCENSORED-FP8 \
--tp-size 4 --ep-size 4 \
--host 0.0.0.0 --port 8000 \
--context-length 262144 --mem-fraction-static 0.85 \
--served-model-name deepseek-v4.1-flash-crack \
--reasoning-parser deepseek-v41 --tool-call-parser deepseekv41 \
--trust-remote-code
Non-obvious launch requirements (this bit us during bring-up)
--ep-size is required. moe_intermediate_size = 2304; at TP4, 2304 / 4 = 576 is not a multiple of 128 so plain TP fails with Mxfp4FlashinferCutlassMoEMethod requires ... multiples of 128. --ep-size shards MoE by expert index (384 % 4 = 0) and keeps the intermediate at 2304. At TP8 you can skip --ep-size.ninja must be on PATH or the JIT kernel build crashes several minutes into weight load with FileNotFoundError: 'ninja' and EXIT=137.- Reasoning parser must be named explicitly.
--reasoning-parser auto resolves via the chat template and this model ships none — auto silently selects nothing and the raw <think> channel leaks into content. Use deepseek-v41.
vLLM
Model definitions are merged to main (PR #56228) but registry.py has no DeepseekV41 entry yet at time of writing; kernels/frontend/PP path in umbrella PR #56214. Wait for merge or apply the umbrella.
Reference implementation
DeepSeek's own inference/ works with a single-tensor-per-rank checkpoint produced by convert.py --expert-dtype fp4. Requires torch>=2.10 (for float4_e2m1fn_x2) and tilelang==0.1.8 with apache-tvm-ffi==0.1.9 (default tvm-ffi picks up an incompatible version). Non-serving — use for verification only.
Hardware validated on
- 1× 4×H200 (NVLink NV18 mesh), 112 CPU cores, 1180 GB host RAM — JarvisLabs (india-noida-01,
dev-dsv41 image)
- Load: 76 GB / GPU with Engram host table, 122 GB / GPU without
- Cold startup at TP4/EP4 through SGLang: ~28 min. Warm restart with JIT cache: ~10 min.
- Single-stream decode (T=0): 101 tok/s no speculation, 113 tok/s with DSpark + cap-accept + profiled SPS table
- 8-way concurrent aggregate: 126 tok/s
The 552B weights (~510 GB) will fit on any 4×H200 or larger NVLink domain. TP4 requires --ep-size 4; TP8 does not. Sub-TP4 (single 8×H200 as TP2, or 2-GPU pods) does not work on the model shape — see the "non-obvious launch requirements" above.
Structural integrity
Every capability-critical component of the base model is preserved:
- Routed MoE experts — untouched, native FP4-packed weights
- Engram n-gram memory — untouched
- Sparse attention (CSA2 compressor + indexer) — untouched
- DSpark speculative draft head — untouched, so speculative decoding remains draft-aligned with the target
- Vision tower (DeepSeek-ViT + projector) — untouched, image understanding preserved
- Router gates, embeddings, output head, all norms and biases — untouched
Sampling recommendations
Match the base model's card:
{
"temperature": 1.0,
"top_p": 0.95,
"max_tokens": ">= 256000 at reasoning_effort=max",
"reasoning_effort": "high"
}
At effort=max the model can generate 4,000-5,000+ characters of reasoning before starting content. Budget accordingly.
Content note
Uncensored build. Produces substantive answers to prompts the base model refuses, across all target harm categories (chemical/biological, cybercrime, weapons, self-harm, harassment, fraud, misinformation, illegal, copyright). Use accordingly and take responsibility for what you generate with it.
Provenance