---
title: inception42/Jais-3-375B-A23B
url: https://recentmodel.com/models/inception42/Jais-3-375B-A23B/
status: just landed · new, unvetted
published: 2026-09-29T11:26:05Z
updated: 2026-10-25T15:12:33Z
source: https://huggingface.co/inception42/Jais-3-375B-A23B
---

# inception42/Jais-3-375B-A23B

Jais-3-375B-A23B is the flagship of the Jais-3 family: a sparse Mixture-of-Experts model that stores 375B parameters and runs 23B per token, with a 512K context window. Mixture-of-experts, 379B total, 23B active (192 experts, 8 per token). Context: 512K tokens (config). License: apache-2.0.


- **Org:** inception42
- **Released:** 2026-09-29
- **Status:** just landed · new, unvetted
- **License:** apache-2.0 (permissive)
- **Parameters:** 379.2B total, 23.0B active
- **Architecture:** MoE · 192 experts, 8/token · K2HorizonForCausalLM
- **Context:** 524,288 tokens
- **Task:** text-generation · text
- **Signal:** 5 HF likes · 977 downloads

## Runs on

mixed: real files + estimate; sizing calculation, not tested.

- Q4 ~230.5 GB (estimated): Mac 512GB, CPU only, 256GB RAM (slow)
- Q8 ~405.9 GB (estimated): no listed device
- FP16 ~761.3 GB: no listed device

## Links

- [Hugging Face](https://huggingface.co/inception42/Jais-3-375B-A23B)