Sep 21, 2026 · Artificial Intelligence

Qwen-Image-2.1: 7B DiT Unifies Text-to-Image, Editing & Transparent Output

Qwen-Image-2.1 releases a 7B-parameter Single-Stream DiT that unifies text-to-image, multi-reference editing up to 10 images, native RGBA transparency, and local region editing, with 33.1GB BF16 weights under a non-commercial research license and a self-reported 60.28 benchmark score.

DiTQwen-Image-2.1RGBA transparency

0 likes · 12 min read

Qwen-Image-2.1: 7B DiT Unifies Text-to-Image, Editing & Transparent Output

Machine Learning Algorithms & Natural Language Processing

Jul 16, 2026 · Artificial Intelligence

Inkling – A 975‑Billion‑Parameter Open‑Weight Multimodal Model from Thinking Machines

Thinking Machines Lab unveiled Inkling, a 975‑billion‑parameter open‑weight multimodal model featuring a hybrid‑expert Transformer, 1‑million‑token context, and extensive benchmark results, alongside the lighter Inkling‑Small, with detailed architecture, training methodology, reinforcement‑learning enhancements, and practical examples of web‑app generation and tool‑calling.

InklingMixture of Expertsbenchmark

0 likes · 15 min read

Inkling – A 975‑Billion‑Parameter Open‑Weight Multimodal Model from Thinking Machines

open-weight model

Qwen-Image-2.1: 7B DiT Unifies Text-to-Image, Editing & Transparent Output

Inkling – A 975‑Billion‑Parameter Open‑Weight Multimodal Model from Thinking Machines