IST Austria Distributed Algorithms and Systems Lab

university

https://ist.ac.at/en/research/alistarh-group/

https://github.com/IST-DASLab

AI & ML interests

None defined yet.

Recent Activity

anm2211 new activity about 2 hours ago

ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF:request for qwen3.8 next

anm2211 new activity about 2 hours ago

ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF:Model outputs garbage in Hermes Agent

anm2211 new activity 2 days ago

ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF:IQ3_S-mtp failed on simple Math

View all activity

Papers

GPTQ-2D: Cubic-Time Two-Sided Adaptive Rounding

MatryoshkaLoRA: Learning Accurate Hierarchical Low-Rank Representations for LLM Fine-Tuning

View all Papers

in ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF about 2 hours ago

request for qwen3.8 next

#3 opened 15 days ago by

Model outputs garbage in Hermes Agent

#35 opened about 3 hours ago by

in ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF 2 days ago

IQ3_S-mtp failed on simple Math

#28 opened 3 days ago by

how to use imatrix-qwen3.8-27b.gguf in llama.cpp

#27 opened 3 days ago by

in ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF 3 days ago

Abliterated PLS!

#15 opened 6 days ago by

Is this model same as Q4KM for coding related tasks?

#22 opened 5 days ago by

in ISTA-DASLab/Qwen3.6-35B-A3B-2Bit-GSQ 4 days ago

GGUF for CPU inference

#2 opened 5 days ago by

in ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF 4 days ago

DFlash2

#23 opened 5 days ago by

#Request: Would u consider moe gguf models at around 8-11gbs size with same quant and mtp overhead attached please?

#18 opened 6 days ago by

authored a paper 4 days ago

GSQ: Highly-Accurate Low-Precision Scalar Quantization for LLMs via Gumbel-Softmax Sampling

Paper • 2604.18556 • Published Apr 20 • 15

in ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF 6 days ago

MLX version Please！

#9 opened 6 days ago by

in ISTA-DASLab/Qwen3.8-27B-3Bit-GSQ 9 days ago

Maybe add vLLM FP8 KVCache calibration scales?

#3 opened 11 days ago by

in ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF 10 days ago

Would you consider GSQ-RCO IQ3_S + MTP for a Qwen3.8-27B coding-focused fine-tune?

#8 opened 10 days ago by

It is recommended to use IQ3_S-MTP for better instruction following and programming capabilities.

#7 opened 12 days ago by

in ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF 11 days ago

Mtp integrated version coming?

#4 opened 14 days ago by

in ISTA-DASLab/Qwen3.8-27B-3Bit-GSQ 11 days ago

Plans to upstream this patch in vLLM?

#2 opened 11 days ago by

in ISTA-DASLab/Qwen3.6-35B-A3B-2Bit-GSQ 11 days ago

can add gguf for us, please?

#1 opened 11 days ago by

updated a model 11 days ago

ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF

Image-Text-to-Text • 27B • Updated 11 days ago • 770k • 961

in ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF 12 days ago

Bigger version?

#2 opened 15 days ago by

Mtp integrated version coming?

#4 opened 14 days ago by