What’s MXFP4? The 4-Bit Secret Powering OpenAI’s GPT‑OSS Models on Modest Hardware
huggingface.co · 944 words · saved by 2 readers
A Blog post by Rakshit Aralimatti on Hugging Face
What’s MXFP4? The 4-Bit Secret Powering OpenAI’s GPT‑OSS Models on Modest Hardware Back to Articles a]:hidden"> What’s MXFP4? The 4-Bit Secret Powering OpenAI’s GPT‑OSS Models on Modest Hardware Community Article Published August 8, 2025 Upvote 36 +30 Rakshit Aralimatti RakshitAralimatti Follow With native MXFP4 precision, GPT‑OSS‑120B runs on a single H100 GPU and GPT‑OSS‑20B fits within just 16 GB of memory. What is MXFP4? The Definition and Its Origin Story MXFP4 stands for Microscaling FP4, a next-generation 4-bit floating-point format created and standardized in early 2024 by the Open Com
saved by
related reading
- 2305.14314arxiv.org
- GPT-4openai.com
- gpt-4.pdfcdn.openai.com
- frontier model training methodologies | Alex Wa's Blogdjdumpling.github.io
- microgptkarpathy.github.io
- GitHub - brexhq/prompt-engineering: Tips and tricks for working with Large Language Models like OpenAI's GPT-4.github.com
- The 4-bitter Lesson | humans&humansand.ai
- 1.5x faster MoE training with custom MXFP8 kernels · Cursorcursor.com
- Transformer Math 101 | EleutherAI Blogblog.eleuther.ai
- GPT in 60 Lines of NumPy | Jay Modyjaykmody.com
- Non-determinism in GPT-4 is caused by Sparse MoE - 152334H152334h.github.io
- GPT-4 Architecture, Infrastructure, Training Dataset, Costs, Vision, MoEsemianalysis.com