🤝 Open to Collab
Sourajit Mandal
Sourajit123
·
AI & ML interests
Self-learning AI, ML and LLMs as a school student purely for fun. 😎
Recent Activity
new activity about 18 hours ago
huggingface/InferenceSupport:Nanbeige/Nanbeige4.2-3B new activity 30 days ago
WeiboAI/VibeThinker-3B:Will data be open sourced? new activity about 1 month ago
Jackrong/GLM-5.1-Reasoning-1M-Cleaned:Will you be making GLM 5.2 as well?Organizations
None yet
Nanbeige/Nanbeige4.2-3B
👍 5
#11183 opened about 18 hours ago
by Sourajit123
Will data be open sourced?
👍 1
3
#15 opened about 1 month ago
by Sourajit123
Will you be making GLM 5.2 as well?
#5 opened about 1 month ago
by Sourajit123
Can you make a huge GLM 5.2 dataset?
3
#2 opened about 1 month ago
by Sourajit123
Can you PLSSS do this with the Qwen3.5 series/release the datasets and training code
1
#3 opened about 1 month ago
by Sourajit123
Waiting for smaller SoTA models and new architectures
👍 2
#8 opened 4 months ago
by Sourajit123
Will the training code and datasets used to open-sourced?
🚀 8
#10 opened 5 months ago
by Sourajit123
Could an EAGLE-3 draft model trained on 1.58bits further speed up LFM2.5 inference?
5
#10 opened 6 months ago
by Sourajit123
How to finetuning these specific datasets with unsloth?
❤️ 3
4
#2 opened 7 months ago
by Sourajit123