2026 KIMI K3
2.8T_OPEN_
SOURCE_LLM.
Тезис: Три блокера при выборе closed-source API: достаточен ли контекст, держит ли coding-бенчмарк в production, когда доступны веса для self-host. 16.07.2026 Moonshot AI (月之暗面) тихо повесила «🎉 Kimi K3 is live!» в API-документации. Этот hardcore-разбор покрывает все опубликованные ключевые точки: спецификация 2,8T, KDA/AttnRes/Stable LatentMoE, полная benchmark-матрица, pricing, 5 путей интеграции, open weights 27.07, scenario matrix, FAQ.
Spec sheet · 30 сек
| Release | 2026-07-16 · silent deploy · zero press event |
| Params | 2.8T · largest OSS model · +75% vs DeepSeek V4 Pro (1.6T) |
| Core | 1M ctx · native vision · MoE 896 experts / 16 active |
| API | Input $3/M · Output $15/M · cache hit $0.30/M |
| OSS | Full weights 2026-07-27 on Hugging Face |
1. Три engineering-вопроса — с цифрами
- Parameter sovereignty после DeepSeek. 18 месяцев потери доли рынка. K3 возвращает титул «largest OSS model» с 2.8T — релиз накануне WAIC 2026. Strategic signal: high.
- 1M ctx ≠ marketing number. Full attention → KV cache O(n²) memory blowup. KDA: -75% KV cache, 6.3× decode speedup at 1M tokens при фиксированном $3/M input.
- «Пишет код» ≠ «пишет код 4+ часа». SWE Marathon: K3 42.0 vs Fable 5 35.0 vs GPT-5.6 Sol 39.0. Ближе к real multi-hour agent sessions, чем one-shot SWE-bench.
2. Kimi K3 — формальное определение
16.07.2026, без keynote: tech blog + pricing page + model ID kimi-k3 — callable immediately.
Kimi K3 — крупнейший open-source LLM по числу параметров: 2.8 trillion (2.8T). +75% vs DeepSeek V4 Pro (1.6T), 2.7× Xiaomi (1.02T), 7×+ Alibaba (397B).
Sparse MoE: 896 experts, 16 active per forward pass (sparsity 1.8%). 1M token context window, native image/video understanding. Target: complex coding, long-doc reasoning, knowledge work.
One-liner: OSS heavyweight coding LLM с native vision и ultra-long memory — на 40% дешевле Claude Opus 4.8, full weights 27 июля.
3. Контекст релиза
- Kimi series: 9 из 12 месяцев на вершине OSS scale leaderboard
- Timing: eve of WAIC 2026 (Shanghai)
- ARR (June 2026): >$300M · round 6 · valuation $31.5B
- API revenue: >70% · international paid users +400%
Closed-source context: Claude Sonnet 5 / GPT-5.6 leak analysis, GPT-5.6 Sol Ultra math proof.
4. Architecture deep-dive
4.1 Kimi Delta Attention (KDA)
Standard Transformer full attention: compute ∝ seq_len². KDA — hybrid linear attention:
- 3:1 ratio linear : full attention layers
- KV cache memory: up to -75%
- At 1M tokens: decode 6.3× faster
- Beats full-attention baseline: short ctx, long ctx, RL scaling
Mental model: full attention = memorize every token. KDA = indexed retrieval + selective full recall on critical positions.
4.2 Attention Residuals (AttnRes)
Selective cross-depth retrieval — model pulls high-value representations from earlier layers without full re-computation. Reported: ~25% training efficiency gain, <2% compute overhead.
4.3 Stable LatentMoE
896 experts, 16 routed per token — 1.8% activation density:
| Component | Function |
|---|---|
| Quantile Balancing | Expert assignment from router score quantiles — zero heuristic hyperparams |
| Per-Head Muon | Per attention-head optimizer — adaptive at scale |
| Sigmoid Tanh Unit (SiTU) | Activation gating control |
| Gated MLA | Multi-head latent attention selectivity |
Net scaling efficiency vs Kimi K2: ~2.5×.
5. Benchmark matrix — raw numbers
| Benchmark | Kimi K3 | Claude Fable 5 | GPT-5.6 Sol | Claude Opus 4.8 | GLM-5.2 |
|---|---|---|---|---|---|
| DeepSWE | 67.5 | 70.0 | 73.0 | 59.0 | 46.2 |
| Program Bench | 77.8 | 76.8 | 77.6 | 71.9 | 63.7 |
| Terminal Bench 2.1 | 88.3 | 84.6 | 88.8 | 84.6 | 82.7 |
| FrontierSWE | 81.2 | 86.6 | 71.3 | 66.7 | 67.3 |
| SWE Marathon | 42.0 | 35.0 | 39.0 | 40.0 | 13.0 |
| BrowseComp | 91.2 | 88.0 | 90.4 | 84.3 | — |
| Automation Bench | 30.8 | 29.1 | 29.7 | 27.2 | 12.9 |
| GPQA-Diamond | 93.5 | 92.6 | 94.1 | 91.0 | 91.2 |
| MMMU-Pro (Vision) | 81.6 | 81.2 | 83.0 | 78.9 | — |
| OmniDocBench | 91.1 | 89.8 | 85.8 | 87.9 | — |
Parse notes:
- SWE Marathon: K3 #1 (42.0) — best proxy for sustained coding sessions
- Program Bench: K3 #1 (77.8) — marginal lead
- FrontierSWE: Fable 5 leads (86.6); K3 (81.2) >> GPT-5.6 Sol (71.3)
- OmniDocBench: K3 #1 (91.1) — vision + long ctx synergy
- Artificial Analysis Intelligence Index v4.1: K3 57.1 (rank 4) — behind Fable 5 (59.9), GPT-5.6 Sol (58.9)
Caveat: Moonshot self-reported. Different inference harnesses (K3: Kimi Code, GPT: Codex, Claude: Claude Code). Independent reproduction pending.
6. Pricing — $/M token breakdown
| Model | Input ($/M) | Output ($/M) | Cache hit | Context |
|---|---|---|---|---|
| Kimi K3 | $3.00 | $15.00 | $0.30 | 1M |
| Claude Sonnet 5 | $3.00 (promo $2) | $15.00 (promo $10) | — | 200K |
| Claude Opus 4.8 | $5.00 | $25.00 | — | 200K |
| GPT-5.5 | $5.00 | $30.00 | — | 400K |
| DeepSeek V4 Pro | $1.74 | $3.48 | $0.145 | 128K |
| Kimi K2.6 | $0.95 | $4.00 | $0.16 | 256K |
- K3 = Sonnet 5 standard ($3/$15), 5× context window
- Cache hit $0.30/M — coding scenarios >90% hit rate
- China API: ¥20/M input · ¥100/M output · cache ¥2/M
- kimi.com free tier; prepaid from ¥199 (promo until Aug 11)
7. Integration paths — 5 routes
- Kimi Web/App: kimi.com — Google OAuth, max reasoning default.
- Official API: key at platform.kimi.ai, OpenAI-compatible:
- OpenRouter:
moonshotai/kimi-k3— official pricing, full 1M ctx. - Cursor multi-route: K3 primary for long-ctx coding; complex repo bugs → Claude Fable 5 fallback.
- Post 27.07: self-host — Hugging Face weights (64+ GPU supernode for production).
8. Scenario routing matrix
| Workload | Route to | Rationale |
|---|---|---|
| Sustained long coding | Kimi K3 | SWE Marathon #1 · longest ctx |
| Complex repo-level bugfix | Claude Fable 5 | FrontierSWE +5.4pt lead |
| Terminal/toolchain agents | GPT-5.6 Sol | Terminal Bench 2.1 leads |
| Ultra-long docs / multimodal | Kimi K3 | OmniDocBench #1 · native vision + 1M |
| Cost-sensitive batch | DeepSeek V4 Pro | Output $3.48/M vs $15/M |
| OSS self-host (near-term) | Kimi K3 (post 27.07) | Strongest downloadable weights |
9. Open weights — 27 July milestone
Moonshot official commitment: 2026-07-27 full model weights on Hugging Face. Post-release K3 becomes:
- Largest downloadable OSS model
- First OSS weights above 2T params
- New community fine-tuning baseline
Quantization-aware training: MXFP4 weights, MXFP8 activations. HF variants: MXFP4/NVFP4. Expected day-1: vLLM, SGLang.
Timeline: Jul 17–20 (WAIC) → Jul 27 (weights drop).
10. FAQ — technical reference
Free tier?
kimi.com free account. API: $3/$15 per 1M tokens.
Local deploy?
Weights Jul 27. Production: 64+ accelerators — not laptop territory.
vs DeepSeek V4 Pro?
K3: 2× params, 1M vs 128K ctx, stronger benchmarks. DeepSeek: $3.48/M output — 4.3× cheaper.
1M ctx useful?
Full codebase ingest, legal/research docs, multi-turn agent memory — flat pricing, no length surcharge.
Low/high reasoning?
Moonshot announced for future updates. Max only today.
11. OSS ecosystem signal
Kimi K3 ≠ parameter vanity project. KDA, AttnRes, Stable LatentMoE are real engineering deltas. On sustained coding, doc understanding, and reasoning — K3 competes with closed-source flagships at reasonable API pricing with full OSS commitment.
Post Jul 27: first 2T+ downloadable baseline for enterprise on-prem, vertical fine-tuning, API-independent agent stacks. Mac devs: K3 API in Cursor for long-ctx coding is the pragmatic path now; track MLX/llama.cpp quant adapters post-weight-drop — see DeepSeek custom chip & inference infra.
12. Verdict: API anywhere, agent stress on remote Mac
kimi.com signup, API key, OpenRouter routing — Windows/Linux sufficient. For Cursor + K3 1M ctx, Xcode agent integration, MLX quant validation — Apple Silicon unified memory + Metal remains lowest-friction path.
Optimal split: local API dev, Kimi Code stress tests, multi-repo SWE Marathon regression, 1M doc batch on MACGPU remote Mac mini M4 — on-demand SSH, agent load isolated from primary machine.