empero-ai / text-generation updated 1 month ago

Qwen3.8-4B-Distill-GGUF

GGUF quantizations of empero-ai/Qwen3.8-4B — a full-parameter distillation of Qwen3.8 2.4T A95B into the Qwen3.5-4B architecture — for llama.cpp, Ollama, LM Studio, Jan, KoboldCpp, and other stock GGUF runtimes.

Params
Context
Downloads 30d
792 K
Likes
134
Commercial use: allowed apache-2.0 Not gated GGUF 1 languages View on Hugging Face ↗

Download history

daily snapshots · 23 days
792 K175 K
Aug 29Sep 5Sep 13Sep 20

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

FileQuantSizeEst. VRAMVerdict on RTX 4090 · 24 GB
Qwen3.8-4B-Q4_K_M.gguf Q4_K_M 2.8 GB 3.6 GB ✅ Runs comfortably
Qwen3.8-4B-Q5_K_M.gguf Q5_K_M 3.2 GB 4.0 GB ✅ Runs comfortably
Qwen3.8-4B-Q6_K.gguf Q6_K 3.6 GB 4.4 GB ✅ Runs comfortably
Qwen3.8-4B-Q8_0.gguf Q8_0 4.6 GB 5.6 GB ✅ Runs comfortably
Qwen3.8-4B-BF16.gguf GGUF 8.7 GB 10.0 GB ✅ Runs comfortably
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub
~ · ollama · Q4_K_M
$ ollama run qwen3-8-4b-distill-gguf

# pin the quantization explicitly
$ ollama run qwen3-8-4b-distill-gguf-q4_k_m
est. VRAM 3.6 GBon RTX 4090 · 24 GBJSON API →

Specifications

Licence
apache-2.0
First seen on the Hub
2026-08-15
Training datasets
undisclosed
Added to our catalog
2026-08-29
Compare with any text-generation model