والله غاكبو ده لاعب عاهة

Why the fuck does @sama need trillions when open source models from china are as good as openai’s models?? And you can fine-tune the open source models. And they are cheaper. And you have more privacy…

I think people still don’t realize how absolutely insane it is that Kimi trained their model with only $4.6 million - and what that actually means. It’s mind-blowing and an incredible achievement by the team.

Kimi-K2 ranks No. 2 on @ArtificialAnlys - used $4.6m to train, open sourced, free for public, super affordable APIs. No 1 has been hyping AGI, raising tens of billions, pivoting to porn, now asking for a taxpayer bailout. Watch their deeds, not words.

Kimi-K2, an open-source model from a small Chinese startup with a fraction of the resources of US tech giants, is now the second most intelligent AI model in the world

Kimi K2 Thinking is the new leading open weights model: it demonstrates particular strength in agentic contexts but is very verbose, generating the most tokens of any model in completing our Intelligence Index evals @Kimi_Moonshot's Kimi K2 Thinking achieves a 67 in the Artificial Analysis Intelligence Index. This positions it clearly above all other open weights models, including the recently released MiniMax-M2 and DeepSeek-V3.2-Exp, and second only to GPT-5 amongst proprietary models. It used the highest number of tokens ever across the evals in Artificial Analysis Intelligence Index (140M), but with MoonShot’s official API pricing of $0.6/$2.5 per million input/output tokens (for the base endpoint), overall Cost to Run Artificial Analysis Intelligence Index comes in cheaper than leading frontier models at $356. Moonshot also offers a faster turbo endpoint priced at $1.15/$8 (driving a Cost to Run Artificial Analysis Intelligence Index result of $1172 for the turbo endpoint - second only to Grok 4 as the most expensive model). The base endpoint is very slow at ~8 output tokens/s while the turbo is somewhat faster at ~50 output tokens/s. The model is one of the largest open weights models ever at 1T total parameters with 32B active. K2 Thinking is the first reasoning model release in Moonshot AI’s Kimi K2 model family, following non-reasoning Kimi K2 Instruct models released previously in July and September 2025. Moonshot AI only refers to post-training in their announcement. This release highlights the continued trend of post-training & specifically RL driving gains in performance for reasoning models and in long horizon tasks involving tool calling. Key takeaways: ➤ Details: text only (no image input), 256K context window, natively released in INT4 precision, 1T total with 32B active (~594GB) ➤ New leader in open weights intelligence: Kimi K2 Thinking achieves a 67 in the Artificial Analysis Intelligence Index. This is the highest open weights score yet and significantly higher than gpt-oss-120b (61), MiniMax-M2 (61), Qwen 235B A22B 2507 (57) and DeepSeek-V3.2-Exp (57). This release continues the trend of open weights models closely following proprietary models in intelligence achieved ➤ China takes back the open weights frontier: Releases from China based AI labs have led in open weights intelligence offered for most of the past year. OpenAI’s gpt-oss-120b release in August 2025 briefly took back the leadership position for the US. Moonshot AI’s K2 Thinking takes back the leading open weights model mantle for China based AI labs ➤ Strong agentic performance: Kimi K2 Thinking demonstrates particular strength in agentic contexts, as showcased by its #2 position in the Artificial Analysis Agentic Index - where it is second only to GPT-5. This is mostly driven by K2 Thinking achieving 93% in 𝜏²-Bench Telecom, an agentic tool use benchmark where the model acts as a customer service agent. This is the highest score we have independently measured. Tool use in long horizon agentic contexts was a strength of Kimi K2 Instruct and it appears this new Thinking variant makes substantial gains ➤ Top open weights coding model, but behind proprietary models: K2 Thinking does not score a win in any of our coding evals - it lands in 6th place in Terminal-Bench Hard, 7th place in SciCode and 2nd place in LiveCodeBench. Compared to open weights models, it is in first or first equal for each of these evals - and therefore comes in ahead of previous open weights leader DeepSeek V3.2 in our Artificial Analysis Coding Index ➤ Biggest leap for open weights in Humanity’s Last Exam: K2 Thinking’s strongest results include Humanity’s Last Exam, where we measured a score of 22.3% (no tools) - an all time high for open weights models and coming in only behind GPT-5 and Grok 4 ➤ Verbosity: Kimi K2 Thinking is very verbose - taking 140M total tokens are used to run our Intelligence Index evaluations, ~2.5x the number of tokens used by DeepSeek V3.2 and ~2x compared to GPT-5. This high verbosity drives both higher cost and higher latency, compared to less verbose models. On Mooshot’s base endpoint, K2 Thinking is 2.5x cheaper than GPT-5 (high) but 9x more expensive than DeepSeek V3.2 (Cost to Run Artificial Analysis Intelligence Index) ➤ Reasoning variant of Kimi K2 Instruct: The model, as per its naming, is a reasoning variant of Kimi K2 Instruct. The model has the same architecture and same number of parameters (though different precision) as Kimi K2 Instruct. It continues to only support text inputs and outputs ➤ 1T parameters but INT4 instead of FP8: Unlike Moonshot’s prior Kimi K2 Instruct releases that used FP8 precision, this model has been released natively in INT4 precision. Moonshot used quantization aware training in the post-training phase to achieve this. The impact of this is that K2 Thinking is only ~594GB, compared to just over 1TB for K2 Instruct and K2 Instruct 0905 - which translates into efficiency gains for inference and training. A potential reason for INT4 is that pre-Blackwell NVIDIA GPUs do not have support for FP4, making INT4 more suitable for achieving efficiency gains on earlier hardware ➤ Access: The model is available on @huggingface with a modified MIT license. @Kimi_Moonshot is serving an official API (available globally) and third party inference providers are already launching endpoints - including @basetenco, @FireworksAI_HQ, @novita_labs, @parasail_io

WOW 👀 Kimi K2 🔥

Dario is crying now 😂😂😂 @DarioAmodei 😛

thank you Moonshot for releasing the best LLM in the world today as an opensource model & with permissive license people like Dario must be losing their minds right now lol

🚀 Hello, Kimi K2 Thinking! The Open-Source Thinking Agent Model is here. 🔹 SOTA on HLE (44.9%) and BrowseComp (60.2%) 🔹 Executes up to 200 – 300 sequential tool calls without human interference 🔹 Excels in reasoning, agentic search, and coding 🔹 256K context window Built as a thinking agent, K2 Thinking marks our latest efforts in test-time scaling — scaling both thinking tokens and tool-calling turns. K2 Thinking is now live on kimi.com in chat mode, with full agentic mode coming soon. It is also accessible via API. 🔌 API is live: platform.moonshot.ai 🔗 Tech blog: moonshotai.github.io/Kimi-K2… 🔗 Weights & code: huggingface.co/moonshotai

Besides the fact that LLMs are performing better and better, what’s actually far more important is how dramatically and consistently their costs are dropping. On average, costs are falling about 40x per year, an absolutely insane figure. But it’s highly significant, because it means, on the one hand, that AI is finding its way into more and more applications, and on the other, that Sam Altman might be right when he says that someday we’ll be able to run AGI locally on our laptops.

This is the morality of the Zionists #USA

Netanyahu has authorised the killing of 63 Palestinians including a large number of children. This is on top of the daily killing of Palestinians by Israel since the ceasefire was announced. This war was never about the hostages for him. It was about staying out of prison.

> "unplayable framerates" > 140-180FPS Peak first world problem

“Researchers need open source. Developers need open source. Companies around the world — we need open source.” -Jensen Huang #NVIDIAGTC

Security footage shows the moment Israeli occupation soldiers encountered a group of Palestinian youths in the West Bank, with one soldier seen throwing a stun grenade as they passed by. No war, no threat - just Israeli terrorists attacking Palestinian youth because they can.

They're not"anti-Israel" - they're pro-humanity. Something your genocide-endorsing paper wouldn't understand.

@grok do you know when they'll fix that

What’s going on with Twitter’s algorithms? I’m seeing weird and irrelevant stuff that has nothing to do with my interests. Fix it @X

Here is self hosted GLM-4.6-REAP-218B running autonomously spinning up tasks, for 4 hours working on issues, throughout the 4 hour period it had 3 tool call errors. It recovered by itself, none of the tasks/sub-tasks failed and the final result is: 1. fixed some url parsing and ai-sdk errors 2. added more chat features (thread splitting, search) 3. added tests for entire feature, tests look solid, low coverage but better than nothing. Running this at home is very good