selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

Qwen3.8-27B-Uncensored-Cyber

Side-by-side specs, memory needs, licence, runtimes and benchmarks. Add up to 4 models.

--vram
any
--gpus
--quant
--add

On your GPU

  • Qwen3.8-27B-Uncensored-Cyber
Property
Overview
TaskVision
Parameters27.4B
Active parametersdense
Context length256K
Input → output
Released15 Aug 2026
LicenceApache 2.0
Gated downloadno
Memory and speed
Fits your hardwarepick a GPU above≈19 GB
VRAM at Q4≈19 GB
VRAM at Q8≈31 GB
VRAM at FP16≈56 GB
Est. generation speedpick a GPU above—
Popularity
Downloads / 30 days959
Likes42
Runs with
Jan
KoboldCpp
llama.cpp
LM Studio
LocalAI
MLX
Ollama
SGLang
vLLM
Ollama tag—
Architecture
ArchitectureQwen3_5ForConditionalGeneration
Layers64
Attention / KV heads24 / 4
Vocabulary248 320
Weights dtypebfloat16
GGUF availableyes

Keep comparing

Memory figures are estimates for an 8K context at the chosen quantisation (Q4_K_M for LLMs by default). Speeds are rough, memory-bound estimates. Self-reported benchmark scores come from the model authors.