🐻熊狸
@bigeagle_xd
🐰 @mianmoe 和旺崽的🐻爸爸
Honestly blown away by how fast I built the Open Vision Reasoner (OVR) homepage — all thanks to Anycoder + Kimi K2! 🛠️ Anycoder is a free, open-source coding playground built entirely in Gradio — super intuitive and perfect for rapid prototyping. Give it a try 👉…
Roo Code 3.23.15-17 are here! • NEW: Moonshot AI provider - another powerful option for your AI workflows • Mistral embeddings - superior codebase indexing at zero cost • Qwen3-235B support with massive 262K context window Plus 50+ other improvements!…
>>> Qwen3-Coder is here! ✅ We’re releasing Qwen3-Coder-480B-A35B-Instruct, our most powerful open agentic code model to date. This 480B-parameter Mixture-of-Experts model (35B active) natively supports 256K context and scales to 1M context with extrapolation. It achieves…
Great to see that original Kimi-VLs get 100K+ monthly downloads on HF! And hope the new 2506 also more downloaded! P.S. Any providers would like to host this small model? @novita_labs @replicate @GroqInc @hyperbolic_labs Would love to help with thinking-template related issues!
今話題のKIMI K2のTechnical PaperにてSwallow Projectから先日発表した論文が言及されていました！ SwallowProjectは、Openなモデル開発プロジェクトとして研究的な新規性とともに、実際に使えるモデルの開発を目指して、引き続き研究開発を進めていきます！ (次のモデルも開発中です)
Thrilled to see our SwallowProject paper cited in KIMI K2's Technical Report (2.2 Pre-training Data)! 🙏 Thank you for recognizing our work! @Kimi_Moonshot
Just read the fantastic report from @Kimi_Moonshot So excited Kimi integrated Evol-Instruct into their Agentic Data Synthesis. It's surreal to see it endure—from the early days of Chat LLMs, through the Reasoning phase, and now powering the new wave of Agent LLMs. What a…
time to reproduce Kimi K2's agentic data synthesis pipeline in the open
just finished the tech report and pushed to github. good night.
A small update on Qwen3-235B-A22B, but a big improvement on its quality! We thought about this decision for a long time, but we believe that providing better-quality performance is more important than the unification at this moment. We are still continuing our research on hybrid…
Bye Qwen3-235B-A22B, hello Qwen3-235B-A22B-2507! After talking with the community and thinking it through, we decided to stop using hybrid thinking mode. Instead, we’ll train Instruct and Thinking models separately so we can get the best quality possible. Today, we’re releasing…
top4 open models are from China, great job brothers!
🧵Top 10 Open Models by Provider Though proprietary models often top the charts, open models are also paired in battle mode, and ranked on our public leaderboards. Here are the top 10 when stacked by top open model by provider. - #1 Kimi K2 (Modified MIT) @Kimi_Moonshot - #2…
we now use K2 + some in-house agentic envs to fetch and process datasets. AI is feeding its own children.
Excited to share what I have been focusing on this year! Inference-time search to optimize Bayesian surprise pushes us towards long-horizon discovery! Introducing "AutoDS": Autonomous Discovery via Surprisal. "It can not only find the diamond in the rough, but also can rule out…
Great science starts with great questions. 🤔✨ Meet AutoDS—an AI that doesn’t just hunt for answers, it decides which questions are worth asking. 🧵
avante.nvim + kimi-k2 + morph fastapply 这套完美搭配真的给我用爽了😍，又便宜又强大又快又丝滑，我昨天 Vibe Coding 了 9 个小时，才花费我 $4.6！其中 k2 花费 $4.1，morph 花费 $0.5 截图为证：图一是 moonshot 消费，图二是 morph 消费，图三是 wakatime 中的编程时间统计图
I found that the combination of kimi-k2 and morph fastapply is incredibly effective. Since kimi-k2 is a bit slow, using morph fastapply means that when calling the edit_file tool, kimi-k2 only need to return the path and edit_code without returning the lengthy search_code part,…
Any vertical agents are fossil fuel (a.k.a., data) generators for the general agents. Verticle agents will vanish in the long run. Only general agents will prevail. This is just getting started.
ChatGPT agent is ready to introduce itself. openai.com/live
wow
🚨 BREAKING: @Kimi_Moonshot’s Kimi-K2 is now the #1 open model in the Arena! With over 3K community votes, it ranks #5 overall, overtaking DeepSeek as the top open model. Huge congrats to the Moonshot team on this impressive milestone! The leaderboard now features 7 different…
We put Kimi K2 @Kimi_Moonshot to the test on @YouWareAI using actual user queries.The performance is shockingly good, and the cost savings are amazing. Here're 8 test cases from our platform to show you the difference👇
Today, avante.nvim has two updates: 1. Fast apply is supported, using morphllm.com to greatly improve the issue of failure when editing files. 2. Only the moonshot provider is used; you can directly use provider = 'moonshot' for one-click access to kimi/k2 models.