GreenBitAI for Apple Silicon
Collection
Expert-paged for Apple Silicon: a machine runs a model larger than it can hold. One signed binary, OpenAI + Codex + Claude Code on one port. • 7 items • Updated • 1
Haojin Yang
yanghaojin
AI & ML interests
None yet
Recent Activity
reacted to theirpost with 👀 about 5 hours ago
DeepSeek-V4.1-Flash-1017, running locally on a Mac
Our 4-bit paged build is up: https://huggingface.co/GreenBitAI/DeepSeek-V4.1-Flash-1017-4bit-paged
On one M3 Ultra Mac Studio with gbx:
• ~22 tok/s standard decode
• ~44 tok/s on code with DeepSeek's official draft head (1.99×)
• ~33 tok/s on chat with the draft head
Why run 1017: DeepSeek reports 77.0 on DeepSWE v1.1 (up from 74.2) and 42.3 on Terminal-Bench 4.0 (up from 31.2), with shorter outputs (https://api-docs.deepseek.com/news/news261017).
It reads text and images, and plugs straight into Claude Code and Codex; the setup is in the model card. The paged format also loads on Macs with less memory, and on an NVIDIA DGX Spark with experts paged from the SSD;
Engine: https://github.com/GreenBitAI/gbx-lm/releases reacted to theirpost with 🚀 about 5 hours ago
DeepSeek-V4.1-Flash-1017, running locally on a Mac
Our 4-bit paged build is up: https://huggingface.co/GreenBitAI/DeepSeek-V4.1-Flash-1017-4bit-paged
On one M3 Ultra Mac Studio with gbx:
• ~22 tok/s standard decode
• ~44 tok/s on code with DeepSeek's official draft head (1.99×)
• ~33 tok/s on chat with the draft head
Why run 1017: DeepSeek reports 77.0 on DeepSWE v1.1 (up from 74.2) and 42.3 on Terminal-Bench 4.0 (up from 31.2), with shorter outputs (https://api-docs.deepseek.com/news/news261017).
It reads text and images, and plugs straight into Claude Code and Codex; the setup is in the model card. The paged format also loads on Macs with less memory, and on an NVIDIA DGX Spark with experts paged from the SSD;
Engine: https://github.com/GreenBitAI/gbx-lm/releases reacted to theirpost with 🔥 about 5 hours ago
DeepSeek-V4.1-Flash-1017, running locally on a Mac
Our 4-bit paged build is up: https://huggingface.co/GreenBitAI/DeepSeek-V4.1-Flash-1017-4bit-paged
On one M3 Ultra Mac Studio with gbx:
• ~22 tok/s standard decode
• ~44 tok/s on code with DeepSeek's official draft head (1.99×)
• ~33 tok/s on chat with the draft head
Why run 1017: DeepSeek reports 77.0 on DeepSWE v1.1 (up from 74.2) and 42.3 on Terminal-Bench 4.0 (up from 31.2), with shorter outputs (https://api-docs.deepseek.com/news/news261017).
It reads text and images, and plugs straight into Claude Code and Codex; the setup is in the model card. The paged format also loads on Macs with less memory, and on an NVIDIA DGX Spark with experts paged from the SSD;
Engine: https://github.com/GreenBitAI/gbx-lm/releasesOrganizations
upvoted a collection 2 days ago
upvoted an article about 1 year ago
Article
4× Smaller, 2× Faster, 100% Smarts - GreenBitAI's 4-bit Models Redefine AI Efficiency
yanghaojin
• • 3 upvoted an article over 2 years ago
Article
GPU Poor Savior: Revolutionizing Low-Bit Open Source LLMs and Cost-Effective Edge Computing
NicoNico
• • 11