Qwen3-11B — 25% Compressed from Qwen3-14B (English · Chat)
This repository is part of the Efficient and Robust AI System (E-AI) Project by
Vincent-Daniel Yun, which publicly releases compressed large language models. This model
is a compressed edition of Qwen/Qwen3-14B with
10 of 40 transformer layers removed (30 layers remain, ≈11B parameters),
then instruction-tuned for chat so it generates coherent responses at lower memory and latency.
⚠️ Language support — English only. This model is tuned on English data and is
English-focused. Other languages (e.g., Korean, Chinese, Japanese) are not officially
supported and may produce degraded or broken output. For open-domain factual questions,
use retrieval (RAG) — the compressed model can hallucinate specific facts.
About E-AI
Modern AI is powerful but heavy. State-of-the-art models are enormous and their inference is slow — still far from human intuition, and far too slow and unreliable to trust in urgent, high-stakes moments.
Two obstacles stand between today's models and AI we can trust in the field. Individually, each model is too large and too slow to run where it is actually needed. Collectively, when many models or agents work together, a single faulty or adversarial member can quietly derail the whole system. E-AI attacks both — making every model lightweight and fast, and keeping teams of agents reliable even when some of them fail.
I started the E-AI (Efficient-AI) project to build compact yet powerful AI that can assist people in disaster scenarios — responding to dangerous accidents quickly and reliably when every second counts.
Method
The pruning method and the recovery method used to build this model are proprietary,
undisclosed methods created by Vincent-Daniel Yun and are not released. The compressed model
is then instruction-tuned for chat (distilled from the base model). Only the resulting model
is shared here.
Results (measured)
All numbers below were measured by us. PPL is on a 2048-token context (lower is better); downstream tasks and MMLU are 0-shot accuracy via lm-eval-harness (higher is better).
Metric
Qwen3-14B (base)
This model (25%)
PPL · WikiText2 ↓
8.64
23.34
PPL · C4 ↓
13.00
26.31
PPL · PTB ↓
14.79
35.54
ARC-c ↑
0.6024
0.4556
ARC-e ↑
0.8279
0.6894
BoolQ ↑
0.8933
0.6263
COPA ↑
0.9000
0.8000
HellaSwag ↑
0.7881
0.6443
OpenBookQA ↑
0.4620
0.3740
RACE ↑
0.4325
0.3933
RTE ↑
0.7762
0.7545
WinoGrande ↑
0.7317
0.6488
Avg. downstream (9) ↑
0.7127
0.5985
MMLU ↑
0.7729
0.6801
Model family — pick your size
All sizes in this release (click to open each model). Memory is measured peak inference
(fp16 and 4-bit, batch 4 × 2048, single 48 GB GPU).
How little GPU memory each option needs relative to the original dense fp16 model
(lower is better; combine compression with 4-bit for the largest savings).
4-bit (and other) quantization can be used with this model — it is a standard Qwen3
architecture, so bitsandbytes 4-bit / 8-bit loading and other PTQ methods apply on top of
the compression. Verified: this model loads and generates correctly in 4-bit, with peak
inference memory ~11.84 GB (vs 13.9 GB for the dense model in 4-bit, and
33.5 GB for the dense model in fp16).
1import torch
2from transformers import AutoModelForCausalLM, AutoTokenizer
34m = AutoModelForCausalLM.from_pretrained(5"daniel-eai/Qwen3-11B-25pct-Compressed-14B-EN-V1", trust_remote_code=True, dtype=torch.float16, device_map="cuda")6tok = AutoTokenizer.from_pretrained("daniel-eai/Qwen3-11B-25pct-Compressed-14B-EN-V1", trust_remote_code=True)78ids = tok("The capital of France is", return_tensors="pt").to("cuda")9print(tok.decode(m.generate(**ids, max_new_tokens=20)[0]))
trust_remote_code=True is required: the model ships a small custom decoder layer in
modeling_qwen3_recovered.py.
Usage — vLLM
vLLM uses its own model implementations, so the custom decoder layer is loaded via a tiny
plugin (provided in this repo under vllm_plugin/). Install it once, then serve normally:
pip install ./vllm_plugin # from a checkout of this repo's vllm_plugin/ folder
python
1from vllm import LLM, SamplingParams
2llm = LLM(model="daniel-eai/Qwen3-11B-25pct-Compressed-14B-EN-V1", trust_remote_code=True, dtype="float16")3print(llm.generate(["The capital of France is"], SamplingParams(max_tokens=20))[0].outputs[0].text)
Other backends: TGI / SGLang / llama.cpp each use their own model graphs and would need
an analogous custom decoder layer; they are not supported out of the box.
License
Apache-2.0, inherited from the base model Qwen/Qwen3-14B.
Acknowledgements
Sincere thanks to Prof. Sai Praneeth Karimireddy (University of Southern California)
for his invaluable advice and feedback throughout this work, and to Prof. Sunwoo Lee
(Inha University) for his guidance and support. We are also grateful to Alibaba (the
Qwen team) for openly releasing the Qwen3-14B base model that made this work possible.
