SAVRN
Search Contact SAVRN

Open-weight model

Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-oQ6e

by Johannes Uusikuu Johneeee/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-oQ6e

This model was quantized using oQ (oMLX v0.7.0.dev2) mixed-precision quantization.

Parameters26.9B
Context262,144
Weights22.5 GB
License
AccessOpen weights
Monthly Downloads

Runs On

What it takes to serve Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-oQ6e (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

This model was quantized using oQ (oMLX v0.7.0.dev2) mixed-precision quantization.

Excerpt from the card by Johannes Uusikuu.

Configuration

Architecture
Qwen3_5ForConditionalGeneration
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
Stored precision
bfloat16
Model type
qwen3_5

Identity and Version

Repository
Johneeee/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-oQ6e
Publisher
Johannes Uusikuu
Task
Not stated by the source
Modality
Other
Library
mlx
Parameters
26.9B parameters
Languages
mlx, oq
Revision
3555d2e643ace453df6151dcff3ac6afa62529f3
First published
2026-09-18
Last updated
2026-09-18

Files and Weights

15 files, 22.5 GB in total. The weights are 5 files totalling 22.5 GB in safetensors.

Weights5 files · 22.5 GB
Configuration4 files · 219.7 KB
Tokenizer3 files · 19.5 MB
Documentation1 file · 368 B
Other1 file · 17.1 KB
Repository1 file · 1.6 KB
Every file
FileTypeSizeSHA-256
model-00001-of-00005.safetensorsWeights5.0 GB 31be05b158a8
model-00002-of-00005.safetensorsWeights5.0 GB 848130ab4761
model-00003-of-00005.safetensorsWeights5.0 GB 2f514fdd3c11
model-00004-of-00005.safetensorsWeights5.0 GB 23750bccd3eb
model-00005-of-00005.safetensorsWeights2.4 GB 18d560816ad4
config.jsonConfiguration6.2 KB
generation_config.jsonConfiguration213 B
model.safetensors.index.jsonConfiguration182.3 KB
oq_imatrix_report.jsonConfiguration30.9 KB
README.mdDocumentation368 B
chat_template.jinjaOther17.1 KB
.gitattributesRepository1.6 KB
tokenizer.jsonTokenizer12.8 MB 0997f410c57a
tokenizer_config.jsonTokenizer8.7 KB
vocab.jsonTokenizer6.7 MB

License and Download

License
Not stated by the source
Access
Open weights, no gate
Download size
22.5 GB
Download from Johannes Uusikuu

Released by Johannes Uusikuu through its official repository on Hugging Face.

Memory Requirements

PrecisionWeights in memory
As published22.5 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 Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-oQ6e

How much GPU memory does Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-oQ6e 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 Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-oQ6e 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 Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-ULTRA-HERETIC-Uncensored-oQ6e's context length?

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