alanzhuly

5 ms·

1.
Hyperlink: On-device AI agent that searches and summarizes all your local files(hyperlink.nexa.ai)
4 points by alanzhuly 11mo ago|0 comments
2.
Qwen3-VL-4B and 8B runs locally on NPU, GPU, and CPU with one SDK(nexa.ai)
3 points by alanzhuly 1y ago|0 comments
3.
We Ran OpenAI GPT-OSS 20B Locally on a Phone(nexa.ai)
2 points by alanzhuly 1y ago|0 comments
4.
GPT-OSS 20B running on a phone(simonwillison.net)
2 points by alanzhuly 1y ago|0 comments
5.
Matthew McConaughey says he wants a private LLM on Joe Rogan Podcast(twitter.com)
4 points by alanzhuly 1y ago|0 comments
6.
How to unify Gemma and Whisper to build a super fast local voice LLM(nexa.ai)
2 points by alanzhuly 2y ago|0 comments
7.
by alanzhuly 2y ago
Hi! I am from Nexa AI. We just improved Omnivision-968M based on your feedback! Here is a preview in our Hugging Face Space: https://huggingface.co/spaces/NexaAIDev/omnivlm-dpo-demo The updated GGUF and safetensor
8.
What you can do with tiny (1B/3B) LLMs in a local RAG system?(nexa.ai)
1 points by alanzhuly 2y ago|0 comments
9.
by alanzhuly 2y ago
If we can have this running locally on mobile phone that would be pretty cool. Imagine receiving a work document (for example, product requirement documents), and then this turning it into a podcast to play for me while I am driving. I thin
10.
Benchmark GGUF model with ONE line of code(github.com)
6 points by alanzhuly 2y ago|1 comments
11.
by alanzhuly 2y ago
Hi Everyone! We built an open-sourced tool to benchmark GGUF models with a single line of code. GitHub Link: https://github.com/NexaAI/nexa-sdk/tree/main/nexa/eval Motivations: GGUF quantization is
12.
Llama.cpp Now Part of the Nvidia RTX AI Toolkit(developer.nvidia.com)
13 points by alanzhuly 2y ago|1 comments
13.
Small Language Models: Survey, Measurements, and Insights(arxiv.org)
1 points by alanzhuly 2y ago|0 comments
14.
by alanzhuly 2y ago
Thanks for reporting. We are investigating this issue. Could you help submit an issue to our GitHub and provide a screenshot of the terminal (with pip show nexaai)? This could help us reproduce this issue faster. Much appreciated!
15.
by alanzhuly 2y ago
Llama3.2 3B feels a lot better than other models with same size (e.g. Gemma2, Phi3.5-mini models). For anyone looking for a simple way to test Llama3.2 3B locally with UI, Install nexa-sdk( https://github.com/NexaAI/nexa
16.
by alanzhuly 2y ago
For anyone looking for a simple alternative for running local models beyond just text, Nexa AI has built an SDK that supports text, audio (STT, TTS), image generation (e.g., Stable Diffusion), and multimodal models! It also has a model hub
17.
by alanzhuly 2y ago
I like the latest qwen2.5 ( https://nexaai.com/Qwen/Qwen2.5-0.5B-Instruct/gguf-q4_0/read... ). It was just released last week. It is one of the best small langauge models right now according to benchmarks. And
18.
Show HN: We built a knowledge hub for running LLMs on edge devices(github.com)
13 points by alanzhuly 2y ago|0 comments
19.
Join Super AI Agent Hackathon at Stanford, Hosted by HuggingFace and Nexa AI(twitter.com)
12 points by alanzhuly 2y ago|0 comments
20.
by alanzhuly 2y ago
Our models work on both mobile apps and web apps. You can use our model's API, which is flexible and can be integrated with any code.
21.
by alanzhuly 2y ago
Currently, to modify the defined functions in the model, you can provide us with a description of your desired functions, and we can customize (fine-tune) a model for you. However, if your use case is similar to the current functions of our
22.
by alanzhuly 2y ago
Our models can handle diverse user inputs well and have high accuracy in our benchmark results. Feel free to check it out here: https://huggingface.co/NexaAIDev/Octopus-v2/blob/main/androi... It can also
23.
by alanzhuly 2y ago
Our models offer developers faster, cheaper alternative to GPT-4o for implementing function-calling AI agent workflows. Developers can use our model APIs and implement each model's functions to integrate AI agent functionalities into t
24.
Show HN: Use functional tokens for AI agents to simplify app workflows(nexa4ai.com)
80 points by alanzhuly 2y ago|10 comments
25.
Google confirms the leaked Search documents are real(theverge.com)
275 points by alanzhuly 2y ago|75 comments
26.
Recovering 4D World from Monocular Video(huggingface.co)
3 points by alanzhuly 2y ago|0 comments
27.
Privacy-Aware Visual Language Models(arxiv.org)
3 points by alanzhuly 2y ago|0 comments
28.
Transformers Can Do Arithmetic with the Right Embeddings(huggingface.co)
1 points by alanzhuly 2y ago|0 comments
29.
Aya 23: Open Weight Releases to Further Multilingual Progress(huggingface.co)
2 points by alanzhuly 2y ago|0 comments
30.
Feds add nine more incidents to Waymo robotaxi investigation(techcrunch.com)
23 points by alanzhuly 2y ago|6 comments

More ›