Mattias Dürrmeier

mattduerrmeier's activity

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

upvoted a paper 1 day ago

The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity

Paper • 2506.06941 • Published Jun 7, 2025 • 17

liked a Space 11 days ago

Running on CPU Upgrade

Featured

3.2k

The Smol Training Playbook

📚

3.2k

The secrets to building world-class LLMs

upvoted an article 18 days ago

view article

Article

Optimization story: Bloom inference

• 

Narsil

•

Oct 12, 2022

• 9

updated a collection 24 days ago

systems

Collection

5 items • Updated 24 days ago

upvoted a paper 24 days ago

Mirage Persistent Kernel: A Compiler and Runtime for Mega-Kernelizing Tensor Programs

Paper • 2512.22219 • Published Dec 22, 2025 • 1

upvoted 3 papers about 1 month ago

Continuous Thought Machines

Paper • 2505.05522 • Published May 8, 2025 • 15

RePo: Language Models with Context Re-Positioning

Paper • 2512.14391 • Published Dec 16, 2025 • 13

Fast-weight Product Key Memory

Paper • 2601.00671 • Published Jan 2 • 7

New activity in deepseek-ai/DeepSeek-V4-Flash about 1 month ago

How to run deepseek on Ada GPUs？Mine is L20.

5

#25 opened about 1 month ago by

XiaoZaiyi

upvoted a paper about 1 month ago

Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence

Paper • 2604.24954 • Published Apr 27 • 25

liked a model about 1 month ago

deepseek-ai/DeepSeek-V4-Flash

Text Generation • 158B • Updated May 6 • 3.44M • • 1.42k

New activity in deepseek-ai/DeepSeek-V4-Flash about 1 month ago

Should the "index_topk" be updated to 1024 just like the Pro model?

1

#23 opened about 1 month ago by

jfcherng

Questions on MoE Hash Routing

#22 opened about 1 month ago by

mattduerrmeier

liked a model about 1 month ago

deepseek-ai/DeepSeek-V4-Pro

Text Generation • 862B • Updated May 6 • 5.51M • • 4.68k

updated a collection about 2 months ago

systems

Collection

5 items • Updated 24 days ago

upvoted a paper about 2 months ago

FlashDecoding++: Faster Large Language Model Inference on GPUs

Paper • 2311.01282 • Published Nov 2, 2023 • 38

updated a collection about 2 months ago

systems

Collection

5 items • Updated 24 days ago

upvoted a paper about 2 months ago

Liger Kernel: Efficient Triton Kernels for LLM Training

Paper • 2410.10989 • Published Oct 14, 2024 • 3

updated a collection about 2 months ago

systems

Collection

5 items • Updated 24 days ago

upvoted a paper about 2 months ago

FlashFormer: Whole-Model Kernels for Efficient Low-Batch Inference

Paper • 2505.22758 • Published May 28, 2025 • 1