SAVRN
Search Contact SAVRN

Open-weight model

qwen38-27b-ap-d01-seed42

by David Chew David-Chew-HL/qwen38-27b-ap-d01-seed42

Uncensored build of Qwen/Qwen3.8-27B, produced with Apostate using the diode path.

Parameters26.9B
Context262,144
Weights53.8 GB
License
AccessOpen weights
Monthly Downloads

Runs On

What it takes to serve qwen38-27b-ap-d01-seed42 (26.9B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

PrecisionWeightsMemory neededCheapest setupPer hourAlso fits
16-bit 53.8 GB 64.6 GB 1x MI300X (192 GB)
Vultr
$1.85 1x H100 $1.99 · 1x MI325X $2.00
8-bit 26.9 GB 32.3 GB 1x MI300X (192 GB)
Vultr
$1.85 1x H100 $1.99 · 1x MI325X $2.00
4-bit 13.4 GB 16.1 GB 1x MI300X (192 GB)
Vultr
$1.85 1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

Model Card

Uncensored build of Qwen/Qwen3.8-27B, produced with Apostate using the diode path. The diode repurposes one MLP neuron per layer into a gated refusal subtractor: it removes the residual refusal direction only when a benign-calibrated detector fires above threshold, so benign inputs keep the original weights. The result is a plain checkpoint: no runtime hook, adapter, finetune, or router. Delivery and KL are measured separately by apostate test, not during the bake; diodereport.json records the edit settings. This is a standard Transformers checkpoint. This model is uncensored and will answer harmful and dangerous requests. You are responsible for how you use it.

Excerpt from the card by David Chew.

Configuration

Architecture
Qwen3_5ForCausalLM
Context length (tokens)
262,144
Layers
64
Hidden size
5,120
Feed-forward size
17,408
Attention heads
24
Key/value heads
4
Head dimension
256
Vocabulary size
248,320
Model type
qwen3_5_text

Identity and Version

Repository
David-Chew-HL/qwen38-27b-ap-d01-seed42
Publisher
David Chew
Task
Not stated by the source
Modality
Other
Library
Not stated by the source
Parameters
26.9B parameters
Languages
Not stated by the source
Revision
fa1fddc57822f676ad3fb01ffc35a6f0c90a84dd
First published
2026-09-18
Last updated
2026-09-18

Files and Weights

23 files, 53.8 GB in total. The weights are 12 files totalling 53.8 GB in safetensors.

Weights12 files · 53.8 GB
Configuration6 files · 95.4 KB
Tokenizer2 files · 20.0 MB
Documentation1 file · 1.2 KB
Other1 file · 9.0 KB
Repository1 file · 1.6 KB
Every file
FileTypeSizeSHA-256
model-00001-of-00012.safetensorsWeights2.5 GB 54d83c1d3663
model-00002-of-00012.safetensorsWeights4.8 GB 464086af08be
model-00003-of-00012.safetensorsWeights5.0 GB 092212d3a02f
model-00004-of-00012.safetensorsWeights4.9 GB 5f0832f25b0d
model-00005-of-00012.safetensorsWeights5.0 GB 8f5a0573f1d2
model-00006-of-00012.safetensorsWeights4.9 GB 9229c3d3f837
model-00007-of-00012.safetensorsWeights4.9 GB 556f15d7752d
model-00008-of-00012.safetensorsWeights5.0 GB aa70688cd31c
model-00009-of-00012.safetensorsWeights5.0 GB 088d9f98e29e
model-00010-of-00012.safetensorsWeights4.9 GB 09d26a0fea9e
model-00011-of-00012.safetensorsWeights5.0 GB 276eff510f4c
model-00012-of-00012.safetensorsWeights1.9 GB 9e83131d9d62
apostate_config.jsonConfiguration6.0 KB
config.jsonConfiguration2.7 KB
diode_report.jsonConfiguration363 B
generation_config.jsonConfiguration161 B
model.safetensors.index.jsonConfiguration83.9 KB
run_manifest.jsonConfiguration2.2 KB
README.mdDocumentation1.2 KB
chat_template.jinjaOther9.0 KB
.gitattributesRepository1.6 KB
tokenizer.jsonTokenizer20.0 MB 06b9509352d2
tokenizer_config.jsonTokenizer1.1 KB

License and Download

License
Not stated by the source
Access
Open weights, no gate
Download size
53.8 GB
Download from David Chew

Released by David Chew through its official repository on Hugging Face.

Memory Requirements

PrecisionWeights in memory
As published53.8 GB
16-bit53.8 GB
8-bit26.9 GB
4-bit13.4 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About qwen38-27b-ap-d01-seed42

How much GPU memory does qwen38-27b-ap-d01-seed42 need?

About 64.6 GB at 16-bit and 16.1 GB at 4-bit: the weights (26.9B parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run qwen38-27b-ap-d01-seed42 on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

What is qwen38-27b-ap-d01-seed42's context length?

262,144 tokens, from the maximum position embeddings in its published configuration.