AdinaY (Adina Yakefu)

posted an update 13 days ago

Post

2891

Kimi K2 Thinking is now live on the hub 🔥

moonshotai/Kimi-K2-Thinking

✨ 1T MoE for deep reasoning & tool use
✨ Native INT4 quantization = 2× faster inference
✨ 256K context window
✨ Modified MIT license

posted an update 14 days ago

Post

459

Chinese open source AI in October wasn’t about bigger models, it was about real world impact 🔥

https://huggingface.co/collections/zh-ai-community/october-2025-china-open-source-highlights

✨ Vision-Language & OCR wave 🌊
- DeepSeek-OCR : 3B
- PaddleOCR-VL : 0.9B
- Qwen3-VL : 2B / 4B / 8B / 32B /30B-A3B
- Open-Bee: Bee-8B-RL
- http://Z.ai Glyph :10B

OCR is industrializing, the real game now is understanding the (long context) document, not just reading it.

✨ Text generation: scale or innovation?
- MiniMax-M2: 229B
- Antgroup Ling-1T & Ring-1T
- Moonshot Kimi-Linear : linear-attention challenger
- Kwaipilot KAT-Dev

Efficiency is the key.

✨ Any-to-Any & World-Model : one step forward to the real world
- BAAI Emu 3.5
- Antgroup Ming-flash-omni
- HunyuanWorld-Mirror: 3D

Aligning with the “world model” globally

✨ Audio & Speech + Video & Visual: released from entertainment labs to delivery platforms
- SoulX-Podcast TTS
- LongCat-Audio-Codec & LongCat-Video by Meituan delivery paltform
- xiabs DreamOmni 2

Looking forward to what's next 🚀

posted an update 19 days ago

Post

444

Kimi Linear🚀 Hybrid linear attention model from Moonshot AI

https://huggingface.co/collections/moonshotai/kimi-linear-a3b

✨ 48B total/ 3B active - MIT license
✨ Up to 1M context
✨ 84.3 on RULER (128k) with 3.98× speedup
✨ Hybrid KDA + MLA architecture for peak throughput & quality

posted an update 23 days ago

Post

1711

Ming-flash-omni Preview 🚀 Multimodal foundation model from AntGroup

inclusionAI/Ming-flash-omni-Preview

✨ Built on Ling-Flash-2.0: 10B total/6B active
✨ Generative segmentation-as-editing
✨ SOTA contextual & dialect ASR
✨ High-fidelity image generation

posted an update 23 days ago

Post

1792

Glyph 🔥 a framework that scales context length by compressing text into images and processing them with vision–language models, released by Z.ai.

Paper:https://huggingface.co/papers/2510.17800
Model:https://huggingface.co/zai-org/Glyph

✨ Compresses long sequences visually to bypass token limits
✨ Reduces computational and memory costs
✨ Preserves meaning through multimodal encoding
✨ Built on GLM-4.1V-9B-Base

posted an update 28 days ago

Post

2624

HunyuanWorld Mirror🔥a versatile feed forward model for universal 3D world reconstruction by Tencent

tencent/HunyuanWorld-Mirror

✨ Any prior in → 3D world out
✨ Mix camera, intrinsics, depth as priors
✨ Predict point clouds, normals, Gaussians & more in one pass
✨ Unified architecture for all 3D task

posted an update about 1 month ago

Post

665

PaddleOCR VL🔥 0.9B Multilingual VLM by Baidu

PaddlePaddle/PaddleOCR-VL

✨ Ultra-efficient NaViT + ERNIE-4.5 architecture
✨ Supports 109 languages 🤯
✨ Accurately recognizes text, tables, formulas & charts
✨ Fast inference and lightweight for deployment

posted an update about 1 month ago

Post

1797

Bee-8B 🐝 open 8B Multimodal LLM built on high quality data, released by
TencentHunyuan

Paper: Bee: A High-Quality Corpus and Full-Stack Suite to Unlock Advanced Fully Open MLLMs (2510.13795)
Model: https://huggingface.co/collections/Open-Bee/bee-8b-68ecbf10417810d90fbd9995

✨ Trained on Honey-Data-15M, a 15M-sample SFT corpus with dual-level CoT reasoning
✨ Backed by HoneyPipe, a transparent & reproducible open data curation suite

posted an update about 1 month ago

Post

323

Reflection ≠ self-correction

Interesting paper on long-chain reasoning 📑 from Miromind AI.
First Try Matters: Revisiting the Role of Reflection in Reasoning Models (2510.08308)

It dives into how LLMs think 🧠 Most reflections confirm, not fix, and true improvement comes from stronger initial reasoning.

posted an update about 1 month ago

Post

487

Ring-1T🔥 the trillion-parameter thinking model released by Ant group, the company behind Alipay

inclusionAI/Ring-1T

✨ 1T params (50B active)- MIT license
✨ 128K context (YaRN)
✨ RLVR, Icepop, and ASystem make trillion-scale RL stable

posted an update about 1 month ago

Post

511

KAT-Dev-72B-Exp🔥 Kuaishou's ( the company behind Kring AI ) new open model for software engineering

Kwaipilot/KAT-Dev-72B-Exp

✨ 72B - Apache2.0
✨ Redesigned attention kernel & training engine for efficient context-aware RL
✨ 74.6% accuracy on SWE-Bench Verified

posted an update about 1 month ago

Post

4412

At the close of the National Holiday🇨🇳, Antgroup drops a new SoTA model.

Ling-1T 🔥 the trillion-parameter flagship of the Ling 2.0 series.

inclusionAI/Ling-1T

✨1T total / 50B active params per token
✨20T+ reasoning-dense tokens (Evo-CoT)
✨128K context via YaRN
✨FP8 training: 15%+ faster, same precision as BF16
✨Hybrid Syntax-Function-Aesthetics reward for front-end & visual generation

1 reply

·

posted an update about 1 month ago

Post

575

Qwen never stops🤯
Meet Qwen3-VL-30B-A3B-Thinking & Instruct 🔥

Qwen/Qwen3-VL-30B-A3B-Instruct
Qwen/Qwen3-VL-30B-A3B-Thinking

✨256K–1M context for long docs & video
✨32 language OCR & stronger visual recognition
✨“Visual Agent” that operates GUIs