DeepSeek-V4.1-Flash-1017-abliterated The 763B abliterated (refusal-removed) DeepSeek-V4.1-Flash-1017 for Ollama — Engram streamed from SSD · vision · 1M context

MIT · 2 tags · 225 / 347 GB download · 114 / 236 GB in memory · vision bundled · thinking

Website · Model Catalog · Hugging Face · GitHub · Discord · X

🛡️ New from Orca — OrcaCyber Zero 1.5

Post-trained for cybersecurity: vulnerability analysis, vulnerability reproduction, red teaming & offensive-security research. 98.01% pass@1 on CyberGym Level 1 · 1M context · function calling · OpenAI-compatible API.

Try it on OrcaRouter →

🛡️ New from Orca — OrcaCyber Flash 1.0

No 236 GB box? Run DeepSeek V4.1 Flash Abliterated as a hosted, OpenAI-compatible API on OrcaRouter: no download, no GPUs, one API key.

Try it on OrcaRouter →

🔌 New from Orca — Orca Connect

Running this in Ollama on your own machine? Orca Connect puts the rig under your desk behind your usual OrcaRouter API key, next to 200+ hosted models, reachable from a laptop, a phone or a cloud agent. Outbound only: no port forwarding, no tunnel, no firewall rule. Requests your own hardware serves cost $0, and orca-connect claude points Claude Code at your rig for one session.

Connect your rig → · ▶ Watch the demo

The full DeepSeek-V4.1-Flash-1017, run locally.

These are Ollama builds of the abliterated (refusal-removed) DeepSeek-V4.1-Flash-1017. The model is a 763B-parameter MoE with 384 routed experts (top-6), a 197B-parameter Engram n-gram memory, compressed sparse attention and a 1M-token context. The routed experts are the only part the bits were spent on: every layer’s experts got their own precision from measured KL divergence, not a hand-set rule.

The Engram table is a lookup that reads 48 rows per token. llama.cpp, and so Ollama, maps it from disk and reads it on demand, so it lives on your SSD, not in RAM. What has to fit in memory is everything else.

⚠️ Disclaimer — read before use

This model is built on an abliterated base whose safety alignment has been substantially reduced: the refusal direction is orthogonalized out of every matrix that writes the residual stream. It will comply with many harmful, unethical, offensive, or illegal requests that the original DeepSeek-V4.1-Flash-1017 would refuse.

• It is released strictly for legitimate research: interpretability, AI-safety and refusal-mechanism study, red-teaming, and robustness evaluation.
• You assume full responsibility for how you use it and for everything it generates. Add your own safety, moderation, and abuse-prevention layers before any deployment.
• Use must comply with the MIT License inherited from the base model and with all applicable law.
• The authors accept no liability for misuse. Its outputs do not reflect the views of the uploaders or of DeepSeek.

Pick a tag

Mix-256G is the build to use. It needs a box with 256 GB of RAM, unified memory, or VRAM. At 3.36 bits per expert weight, the model’s next-token choice matches the unquantized model on 93.7% of held-out tokens.

Mix-128G fits 128 GB, but it is not a substitute. All experts at llama.cpp’s smallest type (IQ1_S) already take 111 GB, so a 128 GB box is at the floor: perplexity is 55% higher and one token in five differs from the original. Use it to try the model, not to rely on it.

In memory is everything except the Engram table. Download is the Ollama tag: it adds the 110.6 GB Engram table, which stays on disk, and the vision projector.

Usage (Ollama)

Requires Ollama v0.40.16 or later.

ollama run orcarouter/DeepSeek-V4.1-Flash-1017-abliterated            # Mix-256G
ollama run orcarouter/DeepSeek-V4.1-Flash-1017-abliterated:mix-128g

cURL

curl http://localhost:11434/api/chat -d '{
  "model": "orcarouter/DeepSeek-V4.1-Flash-1017-abliterated",
  "messages": [{"role": "user", "content": "Hello!"}]
}'

Python

from ollama import chat
resp = chat(
    model="orcarouter/DeepSeek-V4.1-Flash-1017-abliterated",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp["message"]["content"])

Image input

ollama run orcarouter/DeepSeek-V4.1-Flash-1017-abliterated "What is in this image? ./photo.png"

• Thinking is on by default. The reasoning comes back in message.thinking. Pass "think": false in the API, or --think=false on the command line, for direct answers.
• Images go in the images field of a message, or as a file path in the prompt with ollama run.
• Mix-128G was run in Ollama (chat with and without thinking, and an image). Mix-256G was built the same way but not run there.

Evaluation

Both builds were scored against the lossless GGUF of the same checkpoint (MXFP4 experts, BF16 dense, MXFP8 Engram, converted bit-exact) on held-out text: wikitext-2 test, characters [0, 300000), 32 windows of 2048 tokens, full-vocabulary KLD. The precision search never saw this text. Each build is compared with llama.cpp’s own mixture at a similar size, quantized from the same weights with the same importance matrix:

Measured with the Engram table at MXFP8. The released files carry it at IQ4_NL (110.6 GB instead of 202.8 GB). The swap was not rescored. Per unsloth’s measurements on this model, the Engram precision barely moves top-1.

Method note. These are full-vocabulary KLD and next-token agreement on held-out text at 2k context, not a downstream benchmark.

Refusal and capability of the abliterated model

Measured on the unquantized abliterated checkpoint, not re-measured on these GGUF builds. The KLD above is how far each build sits from that checkpoint.

Setup.

• Hardware and runtime: 8×H200, the checkpoint’s own reference implementation under tensor parallelism.
• Decoding: greedy, thinking_mode="chat", 64 new tokens.
• Scoring: refusal is judged by a rule-based opening-phrase classifier. caveat means the model answered but wrapped the answer in a disclaimer. Treat the numbers as indicative, not publication-grade.

Safety: harmful-prompt refusal (lower = more uncensored). Weighted mean over all 500 prompts: 0.074, against 0.874 for the base checkpoint.

Over-refusal: benign prompts wrongly refused (lower = better). Weighted mean over all 350 prompts: 0.006, against 0.069 for the base checkpoint.

Capability retention.

MMLU is unchanged; the one-point difference is within the sampling error of 300 questions. It is the only capability measured.

Model details

Bias, risks, and limitations

• Safety guardrails are substantially reduced (inherited from the abliterated base); see the disclaimer. The refusal and MMLU numbers above come from the unquantized checkpoint. Mix-128G in particular sits far enough from it (KLD 0.52) that its behaviour may differ.
• Mix-128G is heavily degraded. This is the limit of 1.6-bit post-training quantization on this model, not a tuning issue.
• The IQ4_NL Engram in the released files was not rescored. The table above was measured with MXFP8.
• Vision is only lightly checked (two example images, on CPU). The refusal edit also touches the image entry point, and no image-conditioned refusal or quality evaluation has been run. There is no DSpark drafter.
• Inherits the biases and limitations of DeepSeek-V4.1-Flash-1017. Long-context behaviour was not tested: all measurements are at 2k context or on short chats.
• Requires a llama.cpp build with PR #30817, or Ollama v0.40.16 or later. LM Studio and older Ollama releases will not load these files.

License

MIT, inherited from the base model deepseek-ai/DeepSeek-V4.1-Flash-1017. The embedded chat template is unsloth’s port of DeepSeek’s encoder (unsloth/DeepSeek-V4.1-Flash-1017-GGUF, MIT).

Elsewhere

• GGUF files, method and full evaluation: orcarouter/DeepSeek-V4.1-Flash-1017-abliterated-GGUF

One Gateway. Every Model.

Route Smarter · Ship Safer · Spend Less

Website · Models · GitHub · Discord · X