{"slug": "kimi-k3", "title": "Kimi K3", "summary": "Moonshot AI released Kimi K3, a 2.8-trillion-parameter Mixture-of-Experts model with native 1M context and vision capabilities, claiming it ranks #1 on the Artificial Analysis webdev arena. The model is open-weight but requires 64+ accelerators for self-hosting, with cloud API pricing at $3 per million input tokens.", "body_md": "# Kimi K3\n\nMoE premier2.8T total params, Stable LatentMoE (16 of 896 experts active per token) built on Kimi Delta Attention (hybrid linear attention) + Attention Residuals. Gated MLA, Quantile Balancing routing, Per-Head Muon optimizer, SiTU activation; MXFP4 weights / MXFP8 activations (QAT). Native 1M context, native vision (text + image + video in, text out), always-on thinking mode (reasoning_effort currently \"max\" only), full function/tool calling with MCP, automatic prompt caching (>90% cache-hit on coding). Independent eval: #1 on Artificial Analysis webdev arena (1679, ahead of Claude Fable 5 at 1631) and #4 of 189 on the AA Intelligence Index (57) - comparable to Opus 4.8 / GPT-5.5, behind Fable 5 / GPT-5.6 Sol. Vendor benchmarks (GPQA-Diamond 93.5, HLE 43.5, Terminal Bench 2.1 88.3, FrontierSWE 81.2, MMMU-Pro 81.6) are Moonshot self-reported. Open weights announced (world's first open 3T-class model) but not yet public on HuggingFace as of release day - full weights scheduled by 2026-07-27. Cloud API $3/1M in, $15/1M out (cache-hit input $0.30/1M); self-hosting needs a 64+ accelerator supernode, so cloud-only for nearly everyone despite open weights.\n\n- 2800.0B\n- 1000k\n- other\n- Jul 2026\n\n## Scores\n\n## Or run it in the cloud\n\nNo per-token API provider pricing tracked for Kimi K3 yet.\nFor flagship list prices, see the\n[token counter](/token-counter).\n\n[Detailed API pricing page + JSON endpoint →](/models/kimi-k3/pricing)\n\n## Inference cost over time\n\nData accumulates from the first daily sync - longer ranges populate over time. Prices come from OpenRouter snapshots, not a historical API.", "url": "https://wpnews.pro/news/kimi-k3", "canonical_source": "https://tokenstead.ai/models/kimi-k3", "published_at": "2026-07-16 22:28:22+00:00", "updated_at": "2026-07-16 22:28:25.738690+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-products", "ai-infrastructure"], "entities": ["Moonshot AI", "Kimi K3", "Claude Fable 5", "Opus 4.8", "GPT-5.5", "GPT-5.6 Sol", "HuggingFace", "Artificial Analysis"], "alternates": {"html": "https://wpnews.pro/news/kimi-k3", "markdown": "https://wpnews.pro/news/kimi-k3.md", "text": "https://wpnews.pro/news/kimi-k3.txt", "jsonld": "https://wpnews.pro/news/kimi-k3.jsonld"}}