
Huma Abidi
@humaabidi
#AllThingsAI Mom, Advocate for #Diversity and #inclusion #WomenInTech Senior Director, Artificial Intelligence and Deep Learning @ #Intel
ID: 63354446
06-08-2009 03:21:12
372 Tweet
793 Takipçi
279 Takip Edilen

📢Intel Neural Compressor v2.1 released github.com/intel/neural-c… 🔥1) LLM quantization with enhanced SmoothQuant; 2) Native Keras model quantization; 3) Auto-quant mode; 4) TensorFlow INT8 to ONNX INT8; 5) more ... #oneAPI Huma Abidi @AGRamesh13 Song Han Ella Charlaix

Raja Koduri It's still possible for small/low-precision models to beat full-precision models. Our submission on NeuralChat (fine-tuned on Mistral) with 4-bits quantization outperforms the base Mistral model on LLM leaderboard. Practically people may have different dataset and metrics ...

🙌A nice blog from 11tensors V. Antonopoulos that gives the hands-on practice to build AI Agent: Intel NeuralChat-7B LLM performs very well and can replace Open AI Function Calling to build your AI Agent! 🎯medium.com/11tensors/conn… #iamintel #intelai @intelai Huma Abidi Wei Li

📢When AI meets cybersecurity, see how Intel NeuralChat LLM helps here. Happy to share a nice blog "Harnessing the Intel NeuralChat 7B Model for Advanced Fraud Detection". Congrats Saminu Salisu! 🎯Check out the details: finance.yahoo.com/news/harnessin… #intelai #iamintel Huma Abidi


👨💻2023 is year of open LLMs. Is it time to predict for 2024? DM your thoughts. 📢Re-share the blog from Clémentine Fourrier 🍊: huggingface.co/blog/2023-in-l…, incl. Intel NeuralChat-7B and DPO dataset😀 🤗We hope to contribute more to open-source LLM community in 2024! #iamintel Hugging Face


Small model is capable to show big power in keeping high factual consistency rate! Go, Intel NeuralChat-7B model! Thanks Amr Awadallah 🤖 Ofer Mendelevitch Dr. Deb!

🎯Happy to share that Qdrant vector database has been integrated into #OPEA to support RAG! Thanks to the contributor Anush from @qdrant_ngine. Wei Li Huma Abidi Melissa Evers Arun Gupta and Ke. Go, #OPEA!

💕We love open-source and contributed Intel Neural Compressor (github.com/intel/neural-c…) to ONNX community. Now it's available github.com/onnx/neural-co…, as a quantization tool for ONNX models. 🎯Give a try and share us your feedbacks! Wei Li Huma Abidi Melissa Evers Arun Gupta

🔥Wanted to get the best low-bit LLM? Yes, we released a dedicated low-bit open LLM leaderboard for AIPC: huggingface.co/spaces/Intel/l…, inspired from Hugging Face LLM leaderboard! #intelai #inc #GPTQ #AWQ #GGUF Huma Abidi 吕考考 AK ollama martin_casado Jeremy Howard


Enjoyed being part of the panel with Mustafa Jarrar and others at the GAIN assembly United Airlines Nations, SDAIA, ICESCO In Riyadh today



Kudos to Haihao Shen, 吕考考, and Huma Abidi for Intel Neural Chat 7B making the leaderboard for AI hallucination rates! This dashboard evaluates factual consistency and hallucination rates, crucial for trustworthy AI outputs. Our commitment to innovation is setting new


Pretty impressive vs some bigger models. Nice work Haihao Shen Huma Abidi and the rest of the Intel AI team!


My son Faraz Faraz Abidi built this app that helps #specialNeeds and Elderly be safe and independent!
