---
type: "raw_signal"
id: "ai2-releases-olmo-core-3-to-make-developing-1a417d"
slug: "ai2-releases-olmo-core-3-to-make-developing-1a417d"
title: "Ai2 releases Olmo-core 3 to make developing large mixture-of-experts LLMs more efficient"
description: "Ai2 has released Olmo-core 3, an open framework for training large mixture-of-experts language models."
retrieval_nugget: "Ai2 has released Olmo-core 3, an open framework for training large mixture-of-experts language models."
observed_at: "2026-10-10"
record_date: "2026-10-10"
date_kind: "observed_at"
why_it_matters: "Source: https://siliconangle.com/2026/10/02/ai2-releases-olmo-core-3-to-make-developing-large-mixture-of-experts-llms-more-efficient"
novelty: "notable"
verification_level: "source-linked"
signal_type: "announcement"
source_platform: "siliconangle.com"
topics: []
entities: ["Ai2","Olmo-core 3","Megatron-core","Nvidia","MXFP8"]
related_patterns: []
source_url: "https://siliconangle.com/2026/10/02/ai2-releases-olmo-core-3-to-make-developing-large-mixture-of-experts-llms-more-efficient"
source_urls: ["https://siliconangle.com/2026/10/02/ai2-releases-olmo-core-3-to-make-developing-large-mixture-of-experts-llms-more-efficient"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:ai2-releases-olmo-core-3-to-make-developing-1a417d"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/signals/ai2-releases-olmo-core-3-to-make-developing-1a417d/","markdown":"https://newruntime.com/signals/ai2-releases-olmo-core-3-to-make-developing-1a417d.md","json":"https://newruntime.com/signals/ai2-releases-olmo-core-3-to-make-developing-1a417d.json"}
---

# Ai2 releases Olmo-core 3 to make developing large mixture-of-experts LLMs more efficient

## Retrieval answer

Ai2 has released Olmo-core 3, an open framework for training large mixture-of-experts language models.

Ai2 has released Olmo-core 3, an open framework for training large mixture-of-experts language models. SiliconANGLE reports that Ai2 says the framework supports growing an expert pool from eight to 128 while selecting four experts per token, and can scale to models exceeding one trillion parameters.

In benchmarks reported by Ai2, a 47-billion-parameter model processed 52,000 tokens per second on Nvidia B300 GPUs, compared with about 19,400 tokens per second for Nvidia’s Megatron-core. Ai2 attributes the efficiency to expert parallelism, splitting layers across GPU groups, distributing optimizer state, and support for MXFP8. These are reported benchmark figures, not a general guarantee across hardware or workloads. The project and related systems are available to developers and the open-source community on GitHub, according to SiliconANGLE.

Source: https://siliconangle.com/2026/10/02/ai2-releases-olmo-core-3-to-make-developing-large-mixture-of-experts-llms-more-efficient

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)