---

license: mit

base_model: s-zaizen/DeepSeek-V4.1-Flash-Abliterated

tags:

• deepseek
• deepseek_v41
• abliterated
• uncensored
• gguf
• iq2_xxs
• q2_k
• moe

---

DeepSeek-V4.1-Flash-Abliterated-Q2-GGUF

DwarfStar Q2 GGUF of the community-abliterated DeepSeek V4.1 Flash checkpoint

(s-zaizen/DeepSeek-V4.1-Flash-Abliterated, itself abliterated from

deepseek-ai/DeepSeek-V4.1-Flash with Heretic;

refusal keyword rate reported there as 97/100 → 24/100).

Two files: ...-imatrix.gguf (recommended) and the original

DeepSeek-V4.1-Flash-Abliterated-Q2.gguf bootstrap build (kept for

reference; the imatrix build supersedes it).

Recipe

Converted with deepseek41_quantize.py (DwarfStar ds4 repo) using the

project's C quantizers, --quant q2, imatrix-calibrated routed experts

(16-prompt / 26K-token calibration collected on the abliterated Q2):

• Routed MoE experts: IQ2_XXS gate/up, Q2_K down (imatrix-guided on the

stock recipe; bootstrap here — an imatrix-calibrated v2 may follow)

• Attention, shared experts, projections, output: Q8_0 / F16
• Engram FP8 rows packed losslessly at end of file (read from disk as needed,

never resident)

deepseek41_validate_gguf.py layout audit: PASS (1046 tensors).

Quality vs stock Q2

Target-token NLL on 100 tracked official V4.1 Flash continuations

(deepseek-v4.1-flash-20260910-general, ctx 4096, SSD streaming),

same harness for both files:

| file | mean NLL | median NLL | first-token match | top-1 agreement |

| --- | ---: | ---: | ---: | ---: |

| stock DeepSeek-V4.1-Flash-Q2 | 0.346 | 0.313 | 78/100 | 0.901 |

| bootstrap build | 0.395 | 0.395 | 71/100 | 0.888 |

| imatrix build (this file) | 0.359 | 0.340 | 80/100 | 0.904 |

Essentially at parity with stock: the remaining ~0.01 NLL is plausibly

the ablation itself. First-token match and top-1 agreement beat stock

within noise.

Use

Built for DwarfStar ds4 (docs/MODELS.md

"DeepSeek V4.1 Flash"). It is not a general GGUF: the tensor layout,

quantization mix, and Engram packing are ds4-specific.

./download_model.sh ds41f-q2  # stock weights instead (341 GiB)
./ds4 -m DeepSeek-V4.1-Flash-Abliterated-Q2.gguf --ssd-streaming --ctx 32768

On one 128 GB Mac, use --ssd-streaming (152 GiB main weights; Engram

tables stay on disk). Needs a fast local SSD.

License

Base weights © DeepSeek, MIT (retained via the source checkpoint).

Conversion tooling per the ds4 repo license.