Sasi Kiran's picture

1 137

Sasi Kiran

sasikiran

·

AI & ML interests

Large language models

Recent Activity

liked a model about 1 month ago

nvidia/Cosmos-Reason2-8B

reacted to codelion's post with 🔥 about 1 month ago

Introducing Dhara-70M: A diffusion language model that achieves 3.8x higher throughput than autoregressive models! Key findings from our research on optimal architectures for small language models: → Depth beats width: 32 layers outperforms 12 layers at the same parameter count → Best-in-class factuality: 47.5% on TruthfulQA → 10x training efficiency using WSD (Warmup-Stable-Decay) conversion → Canon layers add only 0.13% parameters but improve reasoning We trained on 1B tokens using the optimal 50-30-20 dataset mix (PDFs + filtered web + educational content), then converted to diffusion with just 100M additional tokens. Blog: https://huggingface.co/blog/codelion/optimal-model-architecture Model: https://huggingface.co/codelion/dhara-70m

liked a model about 1 month ago

zai-org/GLM-TTS

View all activity

Organizations

liked 4 models about 1 month ago

nvidia/Cosmos-Reason2-8B

Image-Text-to-Text • 9B • Updated Dec 19, 2025 • 87.1k • 88

zai-org/GLM-TTS

Text-to-Speech • Updated 17 days ago • 578 • 308

browser-use/bu-30b-a3b-preview

Image-Text-to-Text • 31B • Updated Dec 24, 2025 • 61.5k • 243

ServiceNow-AI/Apriel-1.6-15b-Thinker

Image-Text-to-Text • 15B • Updated Dec 22, 2025 • 2.86k • • 268

liked a model 3 months ago

tlennon-ie/qwen-edit-skin

Image-to-Image • Updated Nov 20, 2025 • 19.2k • • 159

liked a Space 3 months ago

The Smol Training Playbook

The secrets to building world-class LLMs

liked a model 3 months ago

tencent/SongGeneration

Text-to-Audio • Updated Oct 23, 2025 • 364 • 299

liked 4 models 5 months ago

stepfun-ai/Step-Audio-2-mini

Any-to-Any • 8B • Updated Sep 5, 2025 • 936 • 247

bytedance-research/USO

Text-to-Image • Updated Sep 1, 2025 • 228 • 186

Wan-AI/Wan2.2-S2V-14B

Image-to-Video • Updated Sep 17, 2025 • 9.21k • 402

zai-org/glm-4-9b-chat-1m

9B • Updated Jul 10, 2025 • 5.2k • 200

liked 4 models 6 months ago

OmniAvatar/OmniAvatar-14B

Text-to-Video • Updated Aug 6, 2025 • 67 • 106

Qwen/Qwen3-30B-A3B-Instruct-2507-FP8

Text Generation • 31B • Updated Sep 17, 2025 • 389k • 106

Wan-AI/Wan2.2-TI2V-5B

Text-to-Video • Updated Aug 7, 2025 • 5.92k • • 491

AIDC-AI/Ovis-U1-3B

Any-to-Any • 4B • Updated Jul 3, 2025 • 1.95k • 212

liked 2 models 7 months ago

vidore/colqwen-omni-v0.1

Visual Document Retrieval • Updated Jul 17, 2025 • 7.38k • 92

moonshotai/Kimi-K2-Instruct

Text Generation • 1T • Updated Nov 7, 2025 • 177k • • 2.31k

liked a Space 7 months ago

Ovis U1 3B

Demo for multimodal understanding and generation

liked a model 7 months ago

nomic-ai/CodeRankEmbed

0.1B • Updated Jun 24, 2025 • 43.3k • 52

liked a model 9 months ago

ByteDance/DreamO

Updated Jun 24, 2025 • 100