Md Ismail Sojal
0xSojalSec
AI & ML interests
Ai-ML-RL-LLM-MoE-SLMs-MLMs-LAMs-VLMs
Re-searcher/ post-training / reasoning models / RAG /
Recent Activity
upvoted
a
paper
about 23 hours ago
MotionStream: Real-Time Video Generation with Interactive Motion
Controls
upvoted
a
collection
8 days ago
🧠 SmolLM3
liked
a Space
8 days ago
HuggingFaceTB/smol-training-playbook
Organizations
3D World
filtered dataset.
single consumer GPU
smaller model
-
Motif-Technologies/Motif-2.6B
Text Generation • 3B • Updated • 477 • 76 -
InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
Paper • 2508.18265 • Published • 202 -
OpenGVLab/InternVL3_5-241B-A28B
Image-Text-to-Text • 241B • Updated • 6.25k • 131 -
openbmb/MiniCPM-V-4_5
Image-Text-to-Text • 9B • Updated • 42.1k • 1.01k
voiceover
lama
Vision-to-Diagram
3D World
locally
filtered dataset.
Reasoner
single consumer GPU
image-to-image editing
smaller model
-
Motif-Technologies/Motif-2.6B
Text Generation • 3B • Updated • 477 • 76 -
InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
Paper • 2508.18265 • Published • 202 -
OpenGVLab/InternVL3_5-241B-A28B
Image-Text-to-Text • 241B • Updated • 6.25k • 131 -
openbmb/MiniCPM-V-4_5
Image-Text-to-Text • 9B • Updated • 42.1k • 1.01k
tools
voiceover
Chains model
lama