---
title: "Discussion on \"Seven speech models, one H100, no clear winner on news broadcasts\" | Hashnode"
description: "Discussion on \"Seven speech models, one H100, no clear winner on news broadcasts\". The Open ASR Leaderboard has a clear #1 this month: a 30B mixture-of-experts model. Below it sit models a tenth its size. We wanted to know if that order holds on real news.\nWe ran seven speech models"
url: "https://hashnode.com/posts/seven-speech-models-one-h100-no-clear-winner-on-news-broadcasts/6af6c86b69ffc4562868f2eb"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Seven speech models, one H100, no clear winner on news broadcasts](https://xplormitylabs.hashnode.dev/seven-speech-models-one-h100-no-clear-winner-on-news-broadcasts?utm_source=hashnode&utm_medium=feed)

The Open ASR Leaderboard has a clear #1 this month: a 30B mixture-of-experts model. Below it sit models a tenth its size. We wanted to know if that order holds on real news. We ran seven speech models

2d ago · 2 min read · xplormitylabs.hashnode.dev  
Read the article

-   [#speech-recognition](https://hashnode.com/tag/speech-recognition)
-   [#machine-learning-models](https://hashnode.com/tag/machine-learning-models)

## Written by

[Xplormity](https://hashnode.com/@xplormitylabs)

@xplormitylabs

  
[View profile](https://hashnode.com/@xplormitylabs)

Local AI benchmarks, research notes and small tools.

Xplormity is an independent lab. We publish local AI benchmarks, research summaries and small free tools.

## Responses

Markdown supported · ⌘↵ to post

No responses yet.
