Vladimir

vstiff's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in unsloth/Qwen3.8-Flash-Next-GGUF 1 day ago

Qwen3.8-Flash-Next-GGUF Unsloth Dynamic 3.0 Quants

👍 3

1

#88 opened 2 days ago by

Vinyl86

New activity in ISTA-DASLab/DeepSeek-V4.1-Flash-1017-GSQ-RCO-GGUF 1 day ago

Unsloth taking shots at ISTA

🔥👀 11

12

#29 opened 2 days ago by

spanspek

upvoted a collection 1 day ago

Ryzen AI — NPU M1.0 SD Models

Collection

Stable Diffusion models for NPU M1.0 • 10 items • Updated 9 days ago • 1

liked a model 4 days ago

agentionai/Qwen3.8-27B-AP-GGUF

Image-Text-to-Text • 27B • Updated 5 days ago • 887k • 101

upvoted a paper 6 days ago

Qwen-Image-PE: Learning Visual Design for Text-to-Image Prompt Enhancement

Paper • 2610.33472 • Published 9 days ago • 34

upvoted a paper 7 days ago

On-Policy Self-Training: Data-Free Self-Improvement in Large Language Models

Paper • 2610.32770 • Published 10 days ago • 48

upvoted 2 collections 9 days ago

HyLo: Long-Context Aware Upcycling for Hybrid LLMs

Collection

Hybrid MLA + Gated DeltaNet / Mamba-2 models from long-context aware upcycling: up to 32x usable context with over 90% KV-cache reduction. • 14 items • Updated 9 days ago • 3

Ryzen AI — NPU M1.0 LLM Models

Collection

50 items • Updated 9 days ago • 12

New activity in ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF 10 days ago

Q2_0 randomly generates Chinese tokens during English responses

8

#60 opened 10 days ago by

sleepynlp

liked a model 10 days ago

amd/FLUX.1-schnell-amdnpu

Text-to-Image • Updated 10 days ago • 1

upvoted an article 11 days ago

view article

Article

NVIDIA Kumo Relational Outperforms Supervised Machine Learning on Relational Data, No Training Required

• 

nvidia

•

11 days ago

• 37

liked a model 11 days ago

RecursiveSI/flint

Text Generation • 40B • Updated 11 days ago • 26.7k • 1.29k

New activity in ornith-ai/Ornith-1.5-122B-GGUF 12 days ago

MTP model?

2

#2 opened 12 days ago by

LaurentPayot

liked a model 12 days ago

ornith-ai/Ornith-1.5-122B-GGUF

Text Generation • 125B • Updated 12 days ago • 595k • 50

upvoted a paper 13 days ago

BlenderWorld: Benchmarking Visual Agents for 3D Creation from Real-World References

Paper • 2610.23986 • Published 19 days ago • 35

upvoted 2 papers 14 days ago

Towards Looped Models Done Right, Part I: Topology, Input Injection, Recurrent-State Design

Paper • 2610.23036 • Published 17 days ago • 20

Can Visual Generation Expand What Unified Multimodal Models Know?

Paper • 2610.21459 • Published 18 days ago • 37

upvoted a paper 15 days ago

NeMo-DCR: Bit-Exact Delta-Compressed Refit for Scalable Agentic RL at Trillion-Parameter Scale

Paper • 2610.08430 • Published 27 days ago • 29

liked a model 16 days ago

unsloth/Qwen-Image-Layered-GGUF

Image-Text-to-Image • 20B • Updated 3 days ago • 25.4k • 76

reacted to yanghaojin's post with 🚀 16 days ago

view post

Post

3,015

Dear all,
we are happy to share that gbx v0.8.0, our local inference engine for Apple Silicon, runs the 105 GB Qwen3.8-Flash-Next on a 48 GB MacBook Pro by streaming MoE experts from the SSD. When the model fits in memory, it decodes up to 50% faster than omlx, Ollama and mlx-lm on the same Mac, with output identical token for token.
Full write-up: https://medium.com/@semamediadata/running-a-105-gb-qwen-on-a-48-gb-macbook-pro-f116fe493a49
Download gbx: https://github.com/GreenBitAI/gbx-lm/releases
Models: https://huggingface.co/collections/GreenBitAI/greenbitai-for-apple-silicon

See translation

• 
• 
• 
• 
• 

• 
• 
• 2 replies

·

🔥

8

8

🚀

2

2

👀

1

1

+

Reply