←
Kimi Developers99 posts
Kimi Developers avatar

Kimi Developers

@KimiDevs

The official Kimi account for developers building with Kimi Code and the Kimi API.

◷ Joined May 2026▧ 27 Photos and videos
Kimi Developers avatar
Kimi Developers@KimiDevs·Oct 23

Kimi K3.1 Pro is here — a 3T-class open-weight model with a 1M-token context window and major upgrades to coding, 3D creation, and financial research and other agentic capabilities.

Introducing Kimi K3.1 Pro: our most capable 3T-class open-weight model. 🔹 1M-token context, 2.8T total / 104B active params (MoE), Multimodal. 🔹 Full-project coding: games in Unreal, Unity & Godot. 🔹 Excels at parametric 3D modeling in Blender & CAD. 🔹 More reliable at finance work: Accurate modeling with hallucination down by 65%. 🔹 Matching the strongest proprietary models across coding, agentic and knowledge benchmarks. Kimi K3.1 Pro and K3.1 Pro Highspeed (2–3× faster) are now live on Kimi.ai, Kimi Work, Kimi Code, and the Kimi API. Open weights by Nov 4, 2026. 🔗 API: platform.kimi.ai 🔗 Tech blog: kimi.ai/blog/kimi-k3-1-pro

Post attachment
SGLang avatar
⟳ Kimi Developers retweeted
SGLang@sgl_project·Nov 4

Kimi K3.1 Pro replaces dense MLA with sparse attention (DSA). It's much more efficient and easier to serve. SGLang delivers on Day 0: · Up to 33.7K tok/s per GPU (AgentX, GB300) · 450 tok/s single user with DFlash2 + MegaKernel · NVIDIA Vera Rubin / Blackwell / Hopper, AMD, TPU, Trainium, PPU

0:29

Releasing the model weights of Kimi K3.1 Pro: huggingface.co/moonshotai/Ki… We'll keep advancing open intelligence.

Post attachment
vLLM avatar
⟳ Kimi Developers retweeted
vLLM@vllm_project·Nov 4

🎉 Kimi K3.1 Pro from @Kimi_Moonshot lands in vLLM on day 0! A 1M-token context window for long coding sessions, 3D creation and research. It scores 65.7% on Terminal-Bench 4.0 with the Claude Code harness. Give Kimi the long task. vLLM brings Day-0 serving across GPUs and TPUs, plus RL support.

0:38

Releasing the model weights of Kimi K3.1 Pro: huggingface.co/moonshotai/Ki… We'll keep advancing open intelligence.

Post attachment
Kimi Developers avatar
Kimi Developers@KimiDevs·Oct 25

Kimi Code Desktop v1.1.3 By default, the Agent completes tasks on its own. To have it delegate tasks, explicitly request delegation or enable swarm mode. The Agent no longer initiates a swarm on its own; swarm only runs when swarm mode is enabled.

Kimi Developers avatar
Kimi Developers@KimiDevs·Oct 25

Agent swarms now start only when you turn on swarm mode: the agent can no longer launch AgentSwarm by itself, so swarms are no longer available in kimi -p or ACP clients such as Zed.

Kimi Developers avatar
Kimi Developers@KimiDevs·Oct 25

Kimi Code CLI 2.3.2 Handle tasks directly by default instead of proactively launching subagents; request delegation explicitly in your message, AGENTS.md, a skill, or an enabled multi-agent mode.

Kimi Developers avatar
Kimi Developers@KimiDevs·Oct 25

We’ve optimized subagent invocation in Kimi Code CLI and the Kimi Code Desktop app for K3.1 Pro, improving orchestration and collaboration on complex tasks.

Kimi Developers avatar
Kimi Developers@KimiDevs·Oct 16

Congrats to the TokenSpeed team! Building long-context agents with Kimi K3:

Introducing TokenSpeed MLA 2 for NVIDIA Vera Rubin 🚀 Kimi @Kimi_Moonshot K3 decode kernel speedups: up to 3.44× vs. MLA 1 on B200 and up to 2.84× on GB200, combining hardware + software gains. Thanks to @NVIDIAAI for early access and engineering collaboration! 📖 lightseek.org/blog/tokenspee… 👉 github.com/lightseekorg/toke…

Post attachment
Kimi Developers avatar
⟳ Kimi Developers retweeted
Kimi Developers@KimiDevs·Sep 24

Kimi Code Desktop Changelog 1.0.3 🌟 Features Click a local HTML file in the conversation to open it as a real page in the built-in browser. Downloads in the browser's download history can be paused, resumed, canceled, and retried, with speed and time left shown. Valid leading slash commands are highlighted in the Composer, with argument hints shown before you type arguments. Right-click a file name in the conversation to copy its path or reveal it in the folder. New settings toggle for automatic session title generation; turn it off to stop AI naming for new sessions. Free-plan accounts are prompted to upgrade when sending with subscription models.

Kimi Developers avatar
Kimi Developers@KimiDevs·Sep 24

🛠 Bug Fixes Fix a vulnerability that let malicious web pages run arbitrary commands through the local embedded server. Fix long conversations freezing on open, stuttering while streaming, and excessive memory use on mobile. Fix the confirmation dialog spinning forever after deleting a session. Fix the subagent model showing as unconfigured in settings and not taking effect when saved. Fix a series of display issues with Swarm cards and member statuses. Fix forked sessions having their titles overwritten by automatic regeneration. Fix the terminal opening the search panel on Cmd+K and Ctrl+C not copying the selection. Fix the sidebar fighting the work panel when both are open. Fix side chat not showing a failure card on failed requests and side chat entries staying enabled on the empty new-session screen. Fix task notification cards losing stream order inside the turn fold. Fix unstable pinned screenshot dragging on Windows.

Kimi Developers avatar
Kimi Developers@KimiDevs·Sep 24

💡Polish Markdown tables in the conversation now have a rounded frame and expand to the full width. Cron jobs show readable names, schedules, and details in the conversation. Address bar suggestions are highlighted, deduplicated, and distinguished by type. The expanded tool panel shows the full Bash command.

Kimi Developers avatar
Kimi Developers@KimiDevs·Sep 21

With the Kimi Code Desktop you can: 🌟 AI coding you can actually see On the redesigned interface, every step the Agent takes is transparent: tool calls, reasoning, and exactly which files changed in each round. Sensitive actions trigger an approval prompt, and three permission modes — always ask, ask when needed, fully automatic — put you in control. 🌟 A complete workflow from conversation to delivery Planning comes first: every plan is reviewable and open to feedback before work begins. Long tasks can move to the background and you can jump back in anytime. And with tower multi-agent mode (experimental), multiple Agents can push forward in parallel. 🌟 Built-in browser A browser lives in the right-hand panel, with tabs that persist alongside each session and page context shared with the Agent — look up docs, preview results, and verify pages without ever switching windows. 🌟 Point at it, change it Annotate screenshots, highlight text to comment, @-mention files, or drag folders straight into the input box. On web pages, click an element or drag-select a region, add a comment, and send it to the Agent — point at it, and it gets changed. Showing the Agent beats telling the Agent. 🌟 Everything stays in one window Built-in terminal, code diff viewer, file preview, and Markdown rendering — all right inside the app. Any file path mentioned in a chat can be opened with a single click. 🌟 Manage all your projects in one place Multiple workspaces and parallel sessions, with tabs, pinning, batch management, AI-powered naming, and global search — so no matter how many projects you juggle, everything stays organized. 🌟 Sign in with one click via kimi.com / kimi.ai, and check your plan usage anytime. Extend capabilities through the plugin marketplace — or bring your own model providers. kimi.ai/code

Kimi Developers avatar
Kimi Developers@KimiDevs·Sep 21

Meet Kimi Code Desktop Drop it into any dev workflow and get faster, more reliable results — programming tasks done in record time. Manage multiple agents in one focused space, run tasks in parallel, and stay in sync with your agents even on long-horizon work. Now available on MacOS & Windows.

Post attachment
Kimi Developers avatar
Kimi Developers@KimiDevs·Sep 15

Kimi Code New Features 💡 Remote Control is now always on; the experimental KIMI_CODE_EXPERIMENTAL_REMOTE_CONTROL flag has been removed. 💡 Add native clipboard support on Linux X11, so copying from the TUI no longer depends on the terminal's OSC 52 support. 💡 Kimi web: Add a Plugins panel to Settings for browsing the plugin marketplace and installing, enabling, disabling, and removing plugins. See the Changelog for more technical entries. kimi.com/code/docs/en/kimi-c…

Kimi Developers avatar
Kimi Developers@KimiDevs·Aug 28

Kimi Code New Feature: Remote Control Kimi Code 0.39.0 add Remote Control as an experimental feature for accessing a local web session remotely. Enable it with KIMI_CODE_EXPERIMENTAL_REMOTE_CONTROL=1, then run kimi rc, kimi web --remote-control , or /remote-control to start it.

Kimi Developers avatar
Kimi Developers@KimiDevs·Aug 13

💡Add an experimental fullscreen TUI mode. Set the KIMI_CODE_TUI_FULL_SCREEN=1 environment variable to enable it. 💡Support rendering LaTeX math formulas in TUI messages as Unicode formulas. See the changelog on GitHub for more technical entries. github.com/MoonshotAI/kimi-c…