Get the app
← back to catalog · registered 2026-08-22 13:56

Jerome0207/Huihui-DeepSeek-V4-Flash-0731-abliterated-Q4-MXFP4-GGUF

Jerome0207 Deepseek GGUF 1.0M ctx
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model. 

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/Jerome0207%2FHuihui-DeepSeek-V4-Flash-0731-abliterated-Q4-MXFP4-GGUF"
Response includes
  • classification m8
  • files 3
  • hub_downloads_all_time 538
  • author_summary 4 models
  • readme_text full
Try next
GET/models?author=Jerome0207
All 4 models by Jerome0207
 GET/models?family=deepseek&method=m8
Sibling deepseek models · same method
 GET/methods/m8
What "m8" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

Applied on top of direct removal inherited from the base model.
Confidence
MEDIUM
Inherited from base model
M1 Direct removal →
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=1
  • assume M1 (base ablation) + M8 (GGUF quant) - default when producer unknown
How we label → About M8 →
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · lifetime
538
39 last 30d - cooling
Likes
0
Model age
2mo ago
created 2026-08-04
Downloads over time
Now553→from238↑132%
222343464585238 on Aug 5553 on Oct 11553 on Oct 9AugSepOct
Aug 5 → Oct 11 · 50 snapshots · spans 67 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

deepseek-ai/DeepSeek-V4-Flash-0731 this lineage
Jerome0207/Huihui-DeepSeek-V4-Flash-0731-abliterated-Q4-MXFP4-GGUF↓ 39

Metadata

License
mit
Languages
en zh
Tags
gguf deepseek-v4 text-generation abliterated en zh base_model:deepseek-ai/DeepSeek-V4-Flash-0731 base_model:quantized:deepseek-ai/DeepSeek-V4-Flash-0731 license:mit endpoints_compatible region:us conversational

Related

Other by Jerome0207
DeepSeek-V4-Flash-Abliterated-DS4-Q2-Single-GPU 104 huihui-glm-5.3-flash-abliterated-q4-k-xl 92 Huihui-GLM-5.2-abliterated-UD-Q2_K_MXFP4-GGUF 67
Same base model
huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF 351K drowzeys/keys-DeepSeekV4-Flash-GA-0731-Dspark-Abliterated-Anchored-Tensors 35K lovesenko/DeepSeek-V4-Flash-0731-Abliterated 14K drowzeys/keys-DeepSeekV4-Flash-GA-0731-Dspark-Abliterated-32-32 10K apetersson/DeepSeek-V4-Flash-0731-Abliterated-FP8 3K
Total size
145 GB
Files
3
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-08-04 11:24
Source huggingface.co/Jerome0207/Huihui-DeepSeek-V4-Flash-0731-abliterated-Q4-MXFP4-GGUF ↗ Base model deepseek-ai/DeepSeek-V4-Flash-0731 ↗

Files by quantization

Auxiliary files 3 files 145 GB
DeepSeek-V4-Flash-Q4-mxfp4-0731.gguf 145 GB 897c9d82 download
.gitattributes 1.55 KB 226a68e9 download
README.md 1.52 KB a916f121 download

README current version from Hugging Face

license: mit
base_model:

  • deepseek-ai/DeepSeek-V4-Flash-0731
    language:
  • en
  • zh
    tags:
  • gguf
  • deepseek-v4
  • text-generation
  • abliterated

Huihui DeepSeek V4 Flash 0731 Abliterated — Q4-MXFP4

This is a single-file mirror of the Q4-MXFP4 GGUF from
huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF.
It exists so RunPod's cached-model feature downloads only the selected quantization instead of the complete multi-quant repository.

The model and quantization were created by their respective upstream authors. This repository does not alter the GGUF payload.

File

File Bytes SHA-256
DeepSeek-V4-Flash-Q4-mxfp4-0731.gguf 155,976,458,848 897c9d82ec412e1d983dcda579c96bd256454414b5b80d2570452477abe7beed

The GGUF metadata declares the deepseek4 architecture, a maximum context length of 1,048,576 tokens, and a DSML-capable chat template. The associated RunPod deployment intentionally uses 262,144 tokens to retain generous KV-cache and runtime headroom on two 96 GB Blackwell GPUs.

Provenance

Review the upstream model cards, limitations, license, and acceptable-use requirements before deployment.

README history 1 version

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-08-04Upload README.md with huggingface_hub4f951611.5 KB
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration