PulseAugur
LIVE 13:59:42
research · [1 source] ·
0
research

Smol AI releases Llama 3.2 with on-device and multimodal model options

By PulseAugur Editorial · Summary by gemini-2.5-flash-lite from 1 source

Smol AI has released Llama 3.2, a new iteration of their open-source large language model. This release includes smaller, on-device versions with 1 billion and 3 billion parameters, alongside more powerful multimodal models featuring 11 billion and 90 billion parameters. The release also incorporates a "Molmo kicker" from AI2, suggesting enhanced capabilities or features. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

RANK_REASON Release of a new open-source model iteration with varying parameter sizes and multimodal capabilities.

Read on Smol AINews →

COVERAGE [1]

  1. Smol AINews TIER_1 ·

    Llama 3.2: On-device 1B/3B, and Multimodal 11B/90B (with AI2 Molmo kicker)

    **Meta** released **Llama 3.2** with new multimodal versions including **3B** and **20B** vision adapters on a frozen Llama 3.1, showing competitive performance against **Claude Haiku** and **GPT-4o-mini**. **AI2** launched multimodal **Molmo 72B** and **7B** models outperforming…

RELATED TOPICS