# Hugging Face — Introducing Olmo-core 3: Open, scalable training infrastructure for large MoEs

- Company: Hugging Face (huggingface.co)
- Announced: 2026-10-01T15:01:43+00:00
- Category: developer-tool-release
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://huggingface.co/blog/allenai/olmocore3
- Record: https://forck.live/items/15589-introducing-olmo-core-3-open-scalable-training-infrastructure-for-large-moes
- Subject: Platform

Hugging Face released Olmo-core 3, an upgrade to its framework for developing large language models featuring a redesigned open mixture-of-experts (MoE) training system. The system is designed to scale MoE training into the trillion-parameter range while preserving computational efficiency. In one benchmark, increasing the expert pool from 8 to 128 while keeping active parameters per token fixed at about 3.2B resulted in total parameter capacity growing from 4.6B to 47B with training throughput falling by less than 5%.

## Evidence

Verbatim from https://huggingface.co/blog/allenai/olmocore3:

> Olmo-core 3 is designed to scale MoE training into the trillion-parameter range while preserving computational efficiency.

---

Record: https://forck.live/items/15589-introducing-olmo-core-3-open-scalable-training-infrastructure-for-large-moes
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md