Ant Open Source

All things open source at Ant Group. We aim to bring high caliber infrastructure FinTech OSS to the community.

[Ant Open Source @ant_oss]

Ring-2.6-1T is open sourced with🧱Agent-optimized, ⏩adjustable thinking effort, 🧠Deep Thinking features, and worth-mentioned xhigh for complex reasoning. #LLM #opensource
🤗 Hugging Face https://t.co/4LzmeilOBi
📷 ModelScope https://t.co/7MKxDhsMfY
And we are looking forward for feedbacks! 💬 Join the Ant Ling Discord community https://t.co/4sdywnkdBY

[Ant Ling @AntLingAGI]

🚀 Ring-2.6-1T is now open source.

A trillion-scale flagship thinking model built for real-world complex tasks: Agent workflows, coding & engineering, long-horizon tasks, complex reasoning, research, and enterprise automation.

It is designed to move beyond “answering” toward execution: understanding context, planning steps, calling tools, and staying stable across long task chains.

Highlights：
- Advanced agentic workflow support.
- Reasoning effort levels: high for agentic tasks, xhigh for complex reasoning.
- Scalable asynchronous RL via the IcePop algorithm, enabling stable, trillion-scale training for long-horizon agentic RL.

[Ant Open Source @ant_oss]

A new milestone for Ant Design!🎉 https://t.co/BlLyvHpaGi
Ant Design is a complete design ecosystem for enterprise-level products, dedicated to building better user experiences across web, mobile, and beyond
#opensource #ReactJS
🎨Explore Ant Design: https://t.co/G0tcEN1nU2

[Ant Design @AntDesignUI]

Ant Design is now part of the React Foundation Supported Ecosystem 🎉 @reactjs @sethwebster

https://t.co/AkeRKMWUSV

[Ant Open Source @ant_oss]

🥳Announcing a milestone of our open-source ecosystem: AReaL of @TheInclusionAI is now part of the @PyTorch Landscape 🎉 https://t.co/dFaHQu5IWf
AReaL is a modular RL infrastructure that connects foundation model training with downstream agent applications.
Built on a fully asynchronous RL training paradigm, AReaL provides: ✅Modular Architecture✅Plug-and-Play RL as a Service✅Ecosystem Compatibility:
Joining PyTorch Landscape validates our commitment to open, collaborative AI development — and our mission to make RL training accessible to every developer.
#OpenSource #PyTorchFoundation #PyTorch #ReinforcementLearning #LLM
🪐Explore AReaL: https://t.co/O1Bw2KwjZr

[Ant Open Source @ant_oss]

🥳Introducing Ling-2.6-1T, built for the Agentic era, and designed for precise instruct task execution!
Highlights: 📊 Intelligence-Efficiency🛠️ Engineering-Task-Friendly🔍 Production-First

Try out our flagship model:
🤗 Hugging Face: https://t.co/8wi28kosx5
📷 ModelScope: https://t.co/4Ob2G4DuET
#inclusionAI #LLM #OpenSource #AgenticAI

[Ant Ling @AntLingAGI]

Last week, we introduced Ling-2.6-1T. Today, Ling-2.6-1T is officially an open model~ 🤗
1T total parameters · 63B active parameters
We bring values to developers by making it easier to test, deploy, customize, and build.
It is optimized to be "token efficiency" for real production needs:
• Lower token overhead: strong intelligence without long reasoning traces
• Reliable multi-step execution: better instruction, tool, context, and workflow control
• Production-ready deployment: from code generation to bug fixing, with broad agent framework compatibility
A sneak pick into the agentic capability in @opencode

[Ant Open Source @ant_oss]

Designed for 🚀high token efficiency, Ling-2.6-flash is now opensource!
✅ Built for Agent scenarios
✅ Hybrid Linear Architecture
✅ Token-Efficiency Optimization
✅ Intelligence Delivered Blazingly Fast
try out now:
🤖 Hugging Face: https://t.co/Z4dnBiWGU5
⚙️ ModelScope: https://t.co/AA2uY7COkR
#OpenSource #LLM #inlcusionAI #AIAgent

[Ant Ling @AntLingAGI]

Ling-2.6-flash is now officially open-sourced!
A fast, token-efficient Instruct model built for real-world agent workflows.
104B total parameters · 7.4B active parameters
Available in BF16, FP8, and INT4 variants for different deployment needs.
Key strengths:
- Fast generation: 215 tokens/s on Artificial Analysis Output Speed
- High token efficiency: only 15M tokens on the full AA Intelligence Index evaluation
- Real task execution: strong performance across coding, document processing, and lightweight agent workflows
- Improved experience: better Chinese-English switching and smoother compatibility with mainstream coding frameworks

[ModelScope @ModelScope2022]

✨ Inclusion AI's LLaDA2.0-Uni is open-source! A single MoE-based diffusion LLM that unifies visual understanding and image generation — natively, in one model.

Download on ModelScope 👉 https://t.co/3CJDxI8GQ1

Built on a single Mask Token Prediction paradigm, LLaDA2.0-Uni handles:
🖼️ Text-to-image synthesis at 1024×1024, with the option to "think" before drawing
🔍 Visual question answering, captioning, and document understanding on par with dedicated VLMs
✏️ Instruction-driven image editing — single or multi-reference, faithful to original details
🎨 Interleaved text-image reasoning, opening the door to a new class of multimodal chains

Released under Apache 2.0 — paper, code, and weights all open.

📄 https://t.co/4mwDOznE4j
🔗 https://t.co/90FxClxuqQ

[LMSYS Org @lmsysorg]

The DeepSeek V4 garbled output bug in open source inference engine is fixed in SGLang.
To everyone affected over the weekend, sorry for the trouble.

Huge thanks to @Ant_Group for landing the fix PR. It was a cross-company, cross-timezone, sub-48-hour marathon. @ollama and @humansand surfaced it first; @nvidia, @AIatMeta, and @FireworksAI_HQ raised the same signal soon after. @deepseek_ai replied in seconds at every hour. @FireworksAI_HQ stayed up late with us until it shipped. @SemiAnalysis_ and @ollama provided the machines that made the debugging possible. The SGLang team dug in through the weekend.

The real OSS is the friends we made along the way.🫶

[Ant Open Source @ant_oss]

🆕We're open-sourcing DR-Venus — a 4B #deepresearch agent built entirely on open data. Built with two-stage training: SFT with strict cleaning → RL with IGPO turn-level optimization, and Dense credit assignment for 200+ turn trajectories.

What's inside: ✅ Training code & recipes ✅ Model checkpoints (GGUF format for edge deployment) ✅Reproducible paper with our technical details
#inclusionAI #OpenSource #AIAgent
🤗 Models: https://t.co/XTAqaFEZyt
📄 Paper: https://t.co/S4E31ImXu1
🐙 Code: https://t.co/k9XCyiqwJr
#OpenSource #DeepResearch #EdgeAI #AIAgent

[Haoxing Chen @Chenhaoxing249]

After two months of teamwork, we’re excited to share our team’s latest achievement — LLaDA2.0-Uni, InclusionAI’s first multimodal LLaDA.
A unified discrete diffusion LLM built for both understanding and generation across text and images.
Highlights:
● One paradigm for VQA, doc understanding, and image generation
● Efficient inference with a new decoding strategy + 8-step distilled decoder
● Interleaved text-image generation enabled by unified discrete representations
(SGLang support soon)
🤗 Hugging Face: https://t.co/bDiucSDEN7
📷 ModelScope: https://t.co/qnztdVyl7U

[Ant Open Source @ant_oss]

🔧Try out and explore how #LLaDA understanding and generate the world.
📄 Technical Report: https://t.co/DxlLxJ3zMy
🐙 Code: https://t.co/sX0lALW2AA
🤗 Model: https://t.co/n15P5dO7jS

[Ant Open Source @ant_oss]

👋 Introducing LLaDA2.0-Uni — the 💥first unified MoE multimodal model in the LLaDA2.0 series of @TheInclusionAI, designed for native multimodal understanding and generation.
#dLLM #inclusionAI #LLaDA #Multimodal #Opensource #DiffusionModels

1/2 Key features: 🧠Chain-of-Thought Generation
LLaDA2.0-Uni doesn't just generate images — it thinks first. Through reasoning-augmented training, the model performs step-by-step reasoning before visual generation.
Achieves 0.78 on WISE-Bench with thinking mode.

[Ant Open Source @ant_oss]

🔬 Built on three core model designs:
1️⃣ LLaDA2.0 Backbone for Unified Discrete Modelling: Leveraging the MoE dLLM architecture from LLaDA2.0, it formulates both multimodal understanding and generation as a unified block-wise mask prediction paradigm.
2️⃣ Fully Semantic Visual Tokens: Departing from traditional VQ methods that focus on reconstruction, LLaDA2.0-Uni transforms image inputs into purely semantic discrete tokens.
3️⃣ Diffusion Decoder for High-Quality Generation: Leveraging a custom Diffusion Decoder optimized for semantic tokens and few-step distillation, LLaDA2.0-Uni delivers high-fidelity image generation in just 8 steps of ultra-fast inference.

[Robbyant @robbyant_brain]

We’re open-sourcing LingBot-Map — our autoregressive model for streaming 3D reconstruction from a single RGB camera. Real-time camera pose estimation. Real-time 3D scene reconstruction. No specialized hardware required.🚀
#EmbodiedAI #Robotics #3DReconstruction #ComputerVision

[InclusionAI @TheInclusionAI]

Meet LingBot-Map, our autoregressive model for streaming 3D reconstruction! Key insights: Keep SLAM's structural 🧠wisdom, make everything end-to-end learnable, and fully open-sourced.
📊 GitHub: https://t.co/iDutHIKCIP
📖 Paper: https://t.co/p7cdrl7bQm
🤗 HuggingFace: https://t.co/9OBzNUxMt8
#inclusionAI #opensource #LingBot

[Ant Open Source @ant_oss]

Since last year, we've been building something for developers: a data-driven insight to what matters in the AI open-source development space. 🚀 Today we introduce the Q1 2026 Agentic AI Landscape with @TheInclusionAI - your ecosystem navigation map.
What we're releasing today:
✅ 50+ projects mapped — from OpenClaw & Claude-Mem to Aden Hive & Paperclip, covering Coding Agents, Personal Assistants, Orchestration Frameworks
✅ Community insights from 21K+ active developers — extreme power law distribution; indie builders & startups dominate; <10% from big tech

More details👉 https://t.co/z8Pq3vvc9d
#AgenticAI #OpenSource #inclusionAI

[Ant Open Source @ant_oss]

📣Today, @AntGroup AI Security Lab and Tsinghua University open-source ClawAegis, a lightweight security framework for autonomous AI agents like #OpenClaw.

🚒ClawAegis delivers multi-layered defense across the agent’s lifecycle, actively blocking threats such as malicious instruction injection and memory poisoning. It integrates seamlessly as a native plugin.

Building on our recent collaboration to patch critical vulnerabilities, we’re thrilled to continue working with the @openclaw community to shape a safer future for autonomous AI🚀🦞
🔧Try out: https://t.co/HmS741JQI5
#opensource #AISecurity #AIAgent

[Ant Open Source @ant_oss]

Let's advance the #OpenSource ecosystem for spatial perception! 🆕🥳New release: RGB-D dataset with 2M real-world captured samples and 1M high-quality rendered samples. https://t.co/gqh8GUF3a9
#inclusionAI #LingBot

[Robbyant @robbyant_brain]

🚀 Exciting news for the spatial perception community!
📷 For too long, the lack of large-scale, real-world depth datasets has been a major bottleneck. Today, we are open-sourcing the RGB-D dataset built for training our spatial perception model LingBot-Depth — and it's massive. 👇