All release briefs
ListedRepository created

Ornith / Ornith-1.5-122B

Ornith-1.5-122B: text generation, with public weights

Ornith has listed Ornith-1.5-122B for text generation. Its official repository includes model weights and a mit license label.

LanguageDownloadable

Automatic fact brief by Privaj Scout · Brief published Oct 20, 2026 · Sources checked Oct 20, 2026

WHAT IT DOES

The documented input is text and the output is text. This establishes its intended task, not its performance against other models.

Inputs
Text
Outputs
Text
License label
mit
Reported tensors
125.09B

Source facts.

  1. The official task tag identifies text generation.

  2. The repository lists downloadable weights under the mit license label.

  3. Its declared architecture is qwen3_5_moe.

Official model repository and license ↗

Before you run it.

The reported tensor count is 125.09B. At 4 bits per stored parameter, weights alone would occupy about 62.54 GB in decimal units. This is a theoretical storage estimate, not measured VRAM or confirmed 4-bit support. Cache, runtime and auxiliary components need additional memory.

What the evidence shows

This fact brief was assembled automatically from the official model API. Repository creation is not a verified launch date. No independent benchmark, speed measurement, or quality ranking is implied.

Read another briefGemini 4 Argon