# Kimi K3: Moonshot AI's 2.8T-Parameter Open-Weight Frontier Model > Source: > Published: 2026-08-07 15:00:00+00:00 # Kimi K3: Moonshot AI's 2.8T-Parameter Open-Weight Frontier Model Kimi K3 is Moonshot AI's open-weight successor to Kimi K2 — a 2.8T-parameter, 104B-activated MoE model built on a new Kimi Delta Attention (KDA) architecture, with native multimodal vision, a 1,048,576-token context window, and published benchmarks against Claude and GPT-class frontier models. - ⭐ 8176 - Python - Kimi K3 License - Updated 2026-08-08 [1M Context Window LLM 2026: Real Tests](https://dibi8.com/resources/llm-frameworks/1m-context-window-llm-2026-real-test/) • [Open Interpreter: A Codex Fork That Emulates Claude Code and Kimi’s Harness](https://dibi8.com/resources/llm-frameworks/openinterpreter-low-cost-model-coding-agent-2026/) *Kimi K3 — from github.com/MoonshotAI/Kimi-K3* ## What Is Kimi K3? [#](#what-is-kimi-k3) **Kimi K3** is Moonshot AI’s newest open-weight model, and per the README, their most capable to date: a **2.8-trillion-parameter** Mixture-of-Experts model with **104B activated parameters** per token, built on a new **Kimi Delta Attention (KDA)** and **Attention Residuals (AttnRes)** architecture. Moonshot AI describes it as “the world’s first open 3T-class model” — native multimodal (text, image, video-frame understanding), a **1,048,576-token context window**, and released with full weights under a custom **Kimi K3 License**. 🔗 **GitHub**: [https://github.com/MoonshotAI/Kimi-K3](https://github.com/MoonshotAI/Kimi-K3) 🤗 **Weights**: [huggingface.co/moonshotai/Kimi-K3](https://huggingface.co/moonshotai/Kimi-K3) 📄 **Tech report**: linked from the repo as `k3_tech_report.pdf` At **8,100+ GitHub stars**, created July 27, 2026 and pushed as recently as August 6, 2026. ## Architecture: What Changed From K2 [#](#architecture-what-changed-from-k2) | Spec | Value | |---|---| | Total parameters | 2.8T | | Activated parameters | 104B | | Layers | 93 (1 dense + 92 MoE) | | Attention composition | 69 KDA + 24 Gated MLA | | Attention heads | 96 (hidden dim 7168) | | Experts | 896 total, 16 selected/token, 2 shared | | Vocabulary | 160K tokens | | Context length | 1,048,576 tokens | | Vision encoder | MoonViT-V2 (401M params) | | Quantization | Native MXFP4 weights / MXFP8 activations (quantization-aware trained) | Per Moonshot AI, the **Stable LatentMoE** framework activating 16 of 896 experts yields “an approximate 2.5× improvement in overall scaling efficiency” over K2. The quantization detail is worth flagging separately: MXFP4/MXFP8 is trained in **from the SFT stage onward**, not bolted on as a post-hoc compression step — the stated goal being broad hardware compatibility without a separate quantization pass degrading quality. ## Benchmark Highlights (Self-Reported, Max Effort) [#](#benchmark-highlights-self-reported-max-effort) Moonshot AI’s README publishes a large comparison table against Claude Fable 5, Claude Opus 4.8, GPT-5.6 Sol, GPT-5.5, and GLM-5.2. A representative slice — K3 doesn’t sweep every category, and the pattern shifts by benchmark type: | Benchmark | Kimi K3 | Best of the rest | |---|---|---| | BrowseComp (agentic web) | 91.2 | GPT-5.6 Sol 90.4 | | MCPMark-Verified (MCP tool use) | 94.5 | GPT-5.6 Sol / GPT-5.5 tied 92.9 | | Terminal-Bench 2.1 | 88.3 | GPT-5.6 Sol 88.8 | | GPQA Diamond (reasoning) | 93.5 | GPT-5.6 Sol 94.1 | | HLE-Full | 43.5 | Claude Fable 5 53.3 | | CritPt (physics reasoning) | 23.4 | GPT-5.6 Sol 32.3 | | Video-MME (w/ subtitles) | 90.0 | GPT-5.6 Sol 89.5 | | OmniDocBench (document vision) | 91.1 | Claude Fable 5 89.8 | | Harvey Lab-AA (legal) | 94.6 | Claude Fable 5 93.6 | **Read this carefully**: these are Moonshot AI’s own numbers from their tech report, not third-party reproductions. K3 leads on agentic tool-use and several vision/document benchmarks, but trails Claude Fable 5 and GPT-5.6 Sol on some of the hardest pure-reasoning evals (HLE-Full, CritPt). Treat it as “strong, benchmark-dependent,” not “best at everything.” ## Deployment and Model Usage [#](#deployment-and-model-usage) Recommended inference engines, per the README: — published recipes at recipes.vllm.ai[vLLM](https://github.com/vllm-project/vllm)— cookbook at docs.sglang.io[SGLang](https://github.com/sgl-project/sglang)**TokenSpeed**— recipes at lightseek.org A hosted, OpenAI/Anthropic-compatible API is available at platform.kimi.ai (model name `kimi-k3` ). **Thinking is always on.** Reasoning effort is set via a `reasoning_effort` field (`"low"` / `"high"` / `"max"` , default `"max"` ), returned as a separate `reasoning_content` field. The one gotcha worth flagging for anyone integrating this: K3 was trained in **preserved thinking history mode**, meaning multi-turn calls must pass the complete prior assistant message back — `reasoning_content` and `tool_calls` included, not just the final `content` string — or the model loses the thread on follow-up turns. For an agent harness, Moonshot AI points to their own **Kimi Code CLI** — run it in a terminal and switch to K3 with the `/model` command. ## Licensing: What Triggers a Paid Agreement [#](#licensing-what-triggers-a-paid-agreement) The **Kimi K3 License** (custom, same shape as K2’s) is permissive by default — free to use, modify, fine-tune, and redistribute — with two revenue-gated conditions: **“Model as a Service” threshold**: if you (plus affiliates) give third parties API-level control over K3’s inputs/parameters/fine-tuning and your combined revenue exceeds**$20M USD over any 12 consecutive months**, you need a separate commercial agreement with Moonshot AI.** Attribution at scale**: if K3 powers a product with** 100M+ monthly active users**or**$20M+/month revenue**, “Kimi K3” must be prominently displayed in that product’s UI. Both conditions are waived for internal use and for end-user products that merely embed K3’s capabilities without exposing model-level control to third parties. ## License [#](#license) **Kimi K3 License** (custom, permissive with revenue-based commercial terms) — see [LICENSE](https://github.com/MoonshotAI/Kimi-K3/blob/main/LICENSE).