cyrano

cross-posted from: https://lemmy.bestiver.se/post/311664

Comments

Llama 4 Models:
  - Both Llama 4 Scout and Llama 4 Maverick use a Mixture-of-Experts (MoE) design with 17B active parameters each.
  - They are natively multimodal: text + image input, text-only output.
  - Key achievements include industry-leading context lengths, strong coding/reasoning performance, and improved multilingual capabilities.
  - Knowledge cutoff: August 2024.

  Llama 4 Scout:
  - 17B active parameters, 16 experts, 109B total.
  - Fits on a single H100 GPU (INT4-quantized).
  - 10M token context window
  - Outperforms previous Llama releases on multimodal tasks while being more resource-friendly.
  - Employs iRoPE architecture for efficient long-context attention.
  - Tested with up to 8 images per prompt.

  Llama 4 Maverick:
  - 17B active parameters, 128 experts, 400B total.
  - 1M token context window.
  - Not single-GPU; runs on one H100 DGX host or can be distributed for greater efficiency.
  - Outperforms GPT-4o and Gemini 2.0 Flash on coding, reasoning, and multilingual tests at a competitive cost.
  - Maintains strong image understanding and grounded reasoning ability.

  Llama 4 Behemoth (Preview):
  - 288B active parameters, 16 experts, nearly 2T total.
  - Still in training; not yet released.
  - Exceeds GPT-4.5, Claude Sonnet 3.7, and Gemini 2.0 Pro on STEM benchmarks (e.g., MATH-500, GPQA Diamond).
  - Serves as the “teacher” model for Scout and Maverick via co-distillation.

  Misc:
  - MoE Architecture: Only 17B parameters activated per token, reducing inference cost.
  - Native Multimodality: Unified text + vision encoder, pre-trained on large-scale unlabeled data.

https://archive.is/2025.04.04-200922/https://www.ft.com/content/b1eda0a8-6449-4fe5-b8c6-a2afebadc823

Joan Didion and John Gregory Dunne met in the late fifties, when she was working at Vogue and he at Time. They married in 1964, and in 1966 they adopted a baby girl, giving her a name from the Yucatán: Quintana Roo. Together, Didion and Dunne lived out one of the most collaborative literary marriages in American history. Last week, after two years of preparation, the New York Public Library opened the Didion-Dunne archive to the public. Among its three hundred and thirty-six boxes of material is a thick file of typewritten notes by Didion describing her sessions with the psychiatrist Roger MacKinnon, beginning in 1999. Addressed to Dunne, the entries are full of direct quotations and written with the immediacy of fresh recollection. Didion was concerned about Quintana and her struggles with depression and alcoholism, but she was preoccupied, too, with aging, with creative fulfillment, with the complex dynamics of their family. She recorded her thoughts with the cool, forensic clarity she was known for.

https://archive.is/Cricr

Joan Didion and John Gregory Dunne met in the late fifties, when she was working at Vogue and he at Time. They married in 1964, and in 1966 they adopted a baby girl, giving her a name from the Yucatán: Quintana Roo. Together, Didion and Dunne lived out one of the most collaborative literary marriages in American history. Last week, after two years of preparation, the New York Public Library opened the Didion-Dunne archive to the public. Among its three hundred and thirty-six boxes of material is a thick file of typewritten notes by Didion describing her sessions with the psychiatrist Roger MacKinnon, beginning in 1999. Addressed to Dunne, the entries are full of direct quotations and written with the immediacy of fresh recollection. Didion was concerned about Quintana and her struggles with depression and alcoholism, but she was preoccupied, too, with aging, with creative fulfillment, with the complex dynamics of their family. She recorded her thoughts with the cool, forensic clarity she was known for.

https://archive.is/Cricr

https://archive.is/oeBUY

https://archive.is/oeBUY