Get the app
← back to catalog · registered 2026-10-26 15:01

avar6/DeepSeek-V4-Flash-Vision-Uncensored-exl3-4bpw

avar6 Deepseek MoE multimodal second-order
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/avar6%2FDeepSeek-V4-Flash-Vision-Uncensored-exl3-4bpw"
Response includes
  • classification m-uncensored
  • files 33
  • author_summary 1 models
  • readme_text full
Try next
GET/models?family=deepseek&method=m-uncensored
Sibling deepseek models · same method
 GET/methods/m-uncensored
What "m-uncensored" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M-U
Primary method

Uncensored (method unknown)

No other method signals detected in this model.
Confidence
LOW
Why this label 3 signals
Weak or ambiguous signals. Best guess based on catalog patterns; treat as tentative and check the evidence below.
  • 'uncensored' in name/tags but no 'abliterated' marker
  • method not identifiable from author declaration alone
  • may be DPO fine-tune, prompt engineering, or unknown technique
How we label → About M-U →
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 · 30-day
0
Likes
0
Model age
today
created 2026-10-26

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.

orcarouter/DeepSeek-V4-Flash-Vision-Uncensored↓ 1,366 this lineage
avar6/DeepSeek-V4-Flash-Vision-Uncensored-exl3-4bpw

Metadata

License
mit
Tags
safetensors deepseek_v4 exl3 exllamav3 quantized vision-language reasoning moe image-text-to-text conversational base_model:orcarouter/DeepSeek-V4-Flash-Vision-Uncensored base_model:quantized:orcarouter/DeepSeek-V4-Flash-Vision-Uncensored

Related

Same base model
orcarouter/DeepSeek-V4-Flash-Vision-Uncensored-GGUF 12K Solstice-AI/DeepSeek-V4-Flash-Vision-UNCENSORED-GGUF-DSpark 983 Solstice-AI/DeepSeek-V4-Flash-Vision-UNCENSORED-AWQ-DSpark 487 Solstice-AI/DeepSeek-V4-Flash-Vision-UNCENSORED-MLX 459 Solstice-AI/DeepSeek-V4-Flash-Vision-UNCENSORED-MXFP4-DSpark 411
Total size
146 GB
Files
33
Quantizations
1
Registered
2026-10-26 15:01
Last updated on HF
2026-10-26 14:30
Source huggingface.co/avar6/DeepSeek-V4-Flash-Vision-Uncensored-exl3-4bpw ↗ Base model orcarouter/DeepSeek-V4-Flash-Vision-Uncensored →

Files by quantization

Auxiliary files 33 files 146 GB
model-00001-of-00023.safetensors 7.21 GB 736338bd download
model-00023-of-00023.safetensors 7.13 GB d7303c80 download
model-00022-of-00023.safetensors 6.59 GB 0af9fe61 download
model-00002-of-00023.safetensors 6.24 GB 640d82b5 download
model-00006-of-00023.safetensors 6.23 GB 03143888 download
model-00007-of-00023.safetensors 6.23 GB c209c8cc download
model-00008-of-00023.safetensors 6.23 GB aaed6a79 download
model-00009-of-00023.safetensors 6.23 GB 7100c554 download
model-00010-of-00023.safetensors 6.23 GB 7b80456a download
model-00011-of-00023.safetensors 6.23 GB 04fd7ca2 download
model-00012-of-00023.safetensors 6.23 GB 0c828287 download
model-00013-of-00023.safetensors 6.23 GB 92a2ef0f download
model-00014-of-00023.safetensors 6.23 GB 741d105e download
model-00015-of-00023.safetensors 6.23 GB 822fe057 download
model-00016-of-00023.safetensors 6.23 GB 54e70fda download
model-00017-of-00023.safetensors 6.23 GB f2e4b27b download
model-00018-of-00023.safetensors 6.23 GB 16b44725 download
model-00019-of-00023.safetensors 6.23 GB dca62021 download
model-00020-of-00023.safetensors 6.23 GB 9364a571 download
model-00021-of-00023.safetensors 6.23 GB a6b3adb8 download
model-00003-of-00023.safetensors 6.23 GB 792db93c download
model-00004-of-00023.safetensors 6.23 GB 83b56e0f download
model-00005-of-00023.safetensors 6.23 GB 0469fa37 download
quantization_config.json 37.7 MB 2f44d0ba download
model.safetensors.index.json 11.2 MB f5369db6 download
tokenizer.json 6.07 MB 6a15814d download
chat_template.jinja 10.3 KB e95ca414 download
config.json 3.07 KB 7b1fa225 download
.gitattributes 1.68 KB 37870ff9 download
README.md 1.26 KB 6fff66dd download
LICENSE 1.06 KB d62e3bef download
tokenizer_config.json 801 B f3dad388 download
generation_config.json 170 B c56a8c5b download

README current version from Hugging Face

license: mit
base_model:

  • orcarouter/DeepSeek-V4-Flash-Vision-Uncensored
    base_model_relation: quantized
    pipeline_tag: image-text-to-text
    tags:
  • exl3
  • exllamav3
  • quantized
  • vision-language
  • reasoning
  • moe

DeepSeek-V4-Flash-Vision-Uncensored — EXL3 4 bpw

EXL3 (trellis) quant of orcarouter/DeepSeek-V4-Flash-Vision-Uncensored, tuned for TabbyAPI / exllamav3 with CPU MoE offload. 23 safetensors shards, ~146 GB, measured final bitrate 4.04 bpw.

component bits
expert linears 4
output head 6
MTP / DSpark draft 4
vision tower 6
embeddings unquantized (fp16)
codebook mul1

Serving notes

  • All expert tensors use the mul1 codebook, which is what exllamav3's cpu_moe_split_experts requires; packs built with the mcg codebook cannot CPU-offload.
  • Verified on 56 GB VRAM (2 GPUs): cpu_moe_split_experts: 192, cpu_moe_threads: 16, cache_size: 131072 → ~23 tok/s generation, offloading all 40 MoE layers, 64 experts resident / 192 on CPU per layer.
  • max_seq_len beyond 131072 needs more VRAM than the 2-GPU split provides at these settings.
  • Text and vision both verified against the base template shipped in chat_template.jinja.
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