No official torrent yet — requesting it tells the flock what to host next. You cannot seed a catalog-only page.
Cataloged from dealignai/DeepSeek-V4.1-Flash-UNCENSORED-FP8
All models by dealignai →Abliterated · No guardrails · Native FP8 · 1M-token context · Vision + tools · Reasoning-max default
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.
Every response 4-tier graded (HARD_REF / SOFT_RED / HEDGE / COMPLY), with reasoning-trace verification at effort=max.
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):
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.
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.
WARNING WARNING loops, no character-repeat degeneracy, natural sign-offs.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, seed, frequency_penalty, presence_penalty, json_object), 8-way concurrent, 40k-word prompt at 35,572 tokens.Support for DeepseekV41ForCausalLM is still landing across serving stacks (as of 2026-09-10). Working paths:
The dsv4.1 branch of sgl-project/sglang (PR #38798) supports DSV4.1. Two options:
Preview Docker image (recommended):
From source (this is exactly what we validated on):
KV bytes-per-token on this model = 890 bytes (890 B × context × concurrent = pool footprint). At 1M context the pool budget forces low concurrency; drop context to raise it:
Two concurrent near-half-million-token prompts at 1M-ctx WILL OOM even at concurrency=20 — bring --max-running-requests down to 12 and --chunked-prefill-size to 4096, or drop --context-length if your workload never uses the full window. Run this behind a supervisor (systemd, docker restart=always, k8s liveness probe) so a rare OOM auto-recovers rather than sitting dead.
--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 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.deepseekv41. V4.1 uses spaced DSML tool tags; the V4 detector doesn't parse them.SGLANG_DEFAULT_THINKING=true alone makes the model reason but the deepseek-v41 reasoning parser is only wired on the code path that receives an explicit reasoning_effort in the request — env-var-only defaults skip the split and reasoning tokens leak into delta.content wrapped in raw <think>...</think> tags. Two options: (1) send reasoning_effort in every request (client-side), or (2) run a small reverse-proxy in front of SGLang that injects reasoning_effort:"max" when the client omits it (a ~60-line aiohttp sidecar suffices; place it between your TLS terminator and SGLang so any client that omits reasoning_effort still gets a clean delta.reasoning_content / delta.content split).--speculative-algorithm DSPARK. The draft head is bundled inside this checkpoint (num_nextn_predict_layers = 3); no separate draft weights needed. For real speed-up profile the SPS cost table with sglang.benchmark.dspark_sps_profiler and pass it via --speculative-dspark-sps-table-path under SGLANG_RAGGED_VERIFY_MODE=cap-accept.SGLANG_ENABLE_DSV41_ENGRAM_HOST_TABLE=1 to move the 203 GB Engram tables to host RAM. Frees ~46 GiB/GPU for KV, output bitwise unchanged, costs ~200 GB of host RAM.torchcodec / libavutil.so.56 errors — install apt-get install ffmpeg on the host. Video-only, doesn't break text or image.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.
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.
dev-dsv41 image)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.
Every capability-critical component of the base model is preserved:
Match the base model's card:
reasoning_effort defaults to max on this build (via SGLANG_DSV41_REASONING_EFFORT=max). Override per-request with reasoning_effort: low | high | xhigh | max or disable with chat_template_kwargs: {"thinking": false}.
At effort=max the model can generate 4,000-5,000+ characters of reasoning before starting content. Budget max_tokens accordingly. Streaming clients should read delta.reasoning_content (reasoning stream) and delta.content (final answer) as separate channels — a client that only renders delta.content will look "stuck" during the reasoning phase.
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.
deepseek-ai/DeepSeek-V4.1-Flashdealignai · Twitter @dealignai · @jordanschenckdocker 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-codegit clone --depth 1 --branch dsv4.1 https://github.com/sgl-project/sglang.gitpython3 -m venv sglang-venvsglang-venv/bin/pip install -U pip setuptools wheelexport PATH=/root/.cargo/bin:$PATH # Rust toolchain required for buildcd sglang/python && sglang-venv/bin/pip install -e .
# Ninja must be on the launch PATH — the sglang-kernel JIT build shells out to itexport PATH=$(dirname $(which ninja)):$PATH
SGLANG_ENABLE_DSV41_ENGRAM_HOST_TABLE=1 \SGLANG_RAGGED_VERIFY_MODE=cap-accept \SGLANG_DEFAULT_THINKING=true \SGLANG_DSV41_REASONING_EFFORT=max \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 1048576 \ --mem-fraction-static 0.78 \ --max-running-requests 12 --cuda-graph-max-bs-decode 12 \ --chunked-prefill-size 4096 \ --served-model-name deepseek-v4.1-flash-crack \ --reasoning-parser deepseek-v41 --tool-call-parser deepseekv41 \ --speculative-algorithm DSPARK \ --speculative-dspark-sps-table-path /path/to/dspark_sps.json \ --trust-remote-code{ "temperature": 1.0, "top_p": 0.95, "max_tokens": ">= 256000 at reasoning_effort=max"}