Nalin Mehra

nalinm01's activity

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

upvoted 2 papers 7 days ago

VC-Attention: Value Smoothing and Softmax Casting for Low-bit Attention

Paper • 2609.15810 • Published 13 days ago • 50

DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression

Paper • 2609.19969 • Published 10 days ago • 181

upvoted a paper 2 months ago

Kimi K3: Open Frontier Intelligence

Paper • 2607.24653 • Published Jul 27 • 524

updated a collection 2 months ago

face

Collection

2 items • Updated Jul 27

upvoted 2 papers 2 months ago

Self Gradient Forcing: Native Long Video Extrapolation

Paper • 2607.20368 • Published Jul 22 • 37

SANA-Video 2.0: Hybrid Linear Attention with Attention Residuals for Efficient Video Generation

Paper • 2607.21553 • Published Jul 23 • 40

commented a paper 2 months ago

SANA-Video 2.0: Hybrid Linear Attention with Attention Residuals for Efficient Video Generation

Paper • 2607.21553 • Published Jul 23 • 40 •

5

liked 2 Spaces 5 months ago

Running on CPU Upgrade

Agents

Featured

1.47k

Open ASR Leaderboard

🏆

1.47k

Compare speech‑recognition models across datasets and languages

Running

Featured

153

Voxtral Realtime WebGPU

💬

153

Real-time speech transcription, entirely in your browser.

New activity in mistralai/Voxtral-Mini-4B-Realtime-2602 5 months ago

Optimize GPU KV cache memory usage

🔥 3

5

#25 opened 7 months ago by

oleslav

liked a Space 8 months ago

Running on CPU Upgrade

Featured

3.31k

The Smol Training Playbook

📚

3.31k

The secrets to building world-class LLMs