Kimi K3 Review
2.8-Trillion-Parameter Open-Source Record

KDA architecture · 1M token context · Full benchmarks · Pricing · API access · 7/27 open weights

Kimi K3 2.8 trillion parameter open-source model review 2026
On the night of July 16, 2026, Moonshot AI quietly posted "Kimi K3 is live!" at the top of its API docs — no big launch event, yet the release of the world's largest open-source AI model by parameter count (2.8 trillion). If your team is evaluating coding Agents, comparing Claude Fable 5 against GPT-5.6 Sol, or waiting for the strongest open weights to land, this review answers: what K3's KDA/AttnRes/Stable LatentMoE architecture actually solves; how it scores on SWE Marathon, OmniDocBench, and other key benchmarks; and what $3/$15 pricing, four access paths, and the July 27 full-weight open-source release mean in practice.
01

What Is Kimi K3? Core Specs of the 2.8-Trillion-Parameter Open-Source Model

On July 16, 2026, Moonshot AI released Kimi K3 — a 2.8-trillion (2.8T) sparse Mixture-of-Experts (MoE) model that surpasses the previous record holder DeepSeek V4 Pro (1.6T) by nearly 75%, is 2.7x the size of Xiaomi's open model (1.02T), and more than 7x Alibaba's (397B). The quiet rollout contrasts sharply with the scale: one technical blog post, one pricing page, and a model ID you can call immediately — kimi-k3.

K3 uses 896 experts with 16 activated per inference (1.8% sparsity); paired with a 1-million-token ultra-long context (roughly the full text of five novels in one pass) and native vision understanding, it is built for complex coding, long-document reasoning, and knowledge work. In one line: an open-source, natively multimodal coding AI with ultra-long memory, priced 40% below Claude Opus 4.8, with full weights open-sourcing on July 27.

01

Scale shock: After DeepSeek's rise eroded Moonshot's market share, K3 is a strategic counter — the Kimi family held the open-source size ceiling for 9 of the past 12 months.

02

Launch timing: Dropped the night before the 2026 World AI Conference (WAIC), July 17–20, with more announcements expected.

03

Commercial surge: ARR passed $300M by June 2026; a sixth funding round this year at a $31.5B pre-money valuation; API revenue exceeds 70% of total, with overseas paid users up 400%.

04

Long-context pain point: Full attention at 1M tokens causes KV cache memory to explode — K3's KDA mechanism exists specifically to solve this.

05

Coding Agent demand: Long-horizon tasks like SWE Marathon require models to stay coherent over multi-hour sessions — the 1M window plus 90%+ cache hit rate is a core selling point.

ParameterValue
Total parameters2.8 trillion (world's largest open-source model)
ArchitectureKDA + AttnRes + Stable LatentMoE
Active experts16 / 896 (sparse MoE, 1.8%)
Context window1,048,576 tokens (1M)
Input modalitiesText, image, video
Reasoning modeAlways-on max effort (low/high coming in future update)
Open weightsJuly 27, 2026 on Hugging Face
02

Kimi K3 Architecture and Benchmarks: KDA, AttnRes, and Stable LatentMoE

Kimi Delta Attention (KDA) is a hybrid linear attention mechanism: it alternates linear and full attention layers in a 3:1 ratio — three linear layers handle local structure (cheap compute), one full attention layer preserves global information flow. Result: KV cache memory drops by up to 75%, decode speed at 1M tokens improves by up to 6.3x, and it beats pure full-attention baselines on both short and long context and RL scaling scenarios.

Attention Residuals (AttnRes) rework depth-dimension residual connections: standard residuals dilute critical representations from early layers; AttnRes adds selective retrieval so the model can pull high-value representations from earlier layers across depth, delivering roughly 25% training efficiency gain with under 2% extra compute.

Stable LatentMoE enables stable training at extreme sparsity (896 experts / 16 active). Supporting techniques:

TechniqueRole
Quantile BalancingDerives expert allocation directly from router score quantiles, eliminating heuristic hyperparameters
Per-Head MuonIndependent optimization per attention head for more adaptive large-scale training
Sigmoid Tanh Unit (SiTU)Improved activation function control
Gated MLAEnhanced attention selectivity

Combined, these innovations raise K3's overall scaling efficiency by roughly 2.5x versus Kimi K2. Moonshot's self-reported benchmark data below (each model uses its own reasoning harness; independent third-party replication is still underway):

BenchmarkKimi K3Claude Fable 5GPT-5.6 SolClaude Opus 4.8GLM-5.2
DeepSWE67.570.073.059.046.2
Program Bench77.876.877.671.963.7
Terminal Bench 2.188.384.688.884.682.7
FrontierSWE81.286.671.366.767.3
SWE Marathon42.035.039.040.013.0
BrowseComp91.288.090.484.3
Automation Bench30.829.129.727.212.9
GPQA-Diamond93.592.694.191.091.2
MMMU-Pro (vision)81.681.283.078.9
OmniDocBench91.189.885.887.9

SWE Marathon tests sustained long-horizon coding — K3 leads at 42.0 by a wide margin. On Artificial Analysis Intelligence Index v4.1, K3 scores 57.1 (fourth overall), trailing Claude Fable 5 (59.9) and GPT-5.6 Sol (58.9) by just 2.8 points.

03

Kimi K3 Pricing: How Much Does It Cost vs Claude, GPT, and DeepSeek?

K3 standard pricing matches Claude Sonnet 5 ($3/$15 per 1M tokens) but offers a 5x larger context window (1M vs 200K). Cache-hit input drops to $0.30/M (one-tenth of standard); Moonshot reports over 90% cache hit rates in coding workflows, making effective input cost extremely low. OpenRouter's 7-day weighted average confirms effective input cost around $0.55/M.

ModelInput ($/M)Output ($/M)Cache-hit inputContext
Kimi K3$3.00$15.00$0.301M
Claude Sonnet 5$3.00 (promo $2)$15.00 (promo $10)200K
Claude Opus 4.8$5.00$25.00200K
GPT-5.5$5.00$30.00400K
DeepSeek V4 Pro$1.74$3.48$0.145128K
Kimi K2.6$0.95$4.00$0.16256K

Versus Claude Opus 4.8, K3 beats it on multiple benchmarks while input costs 60% and output costs 40% as much. China API pricing: input ¥20/M, output ¥100/M, cache-hit ¥2/M. Consumer tier: free account on Kimi.com; prepaid plans from ¥199 (promo through August 11).

04

How to Use Kimi K3 Now: Four Access Methods and Six-Step Runbook

01

Kimi web/app (simplest): Visit kimi.com, register (Google sign-in supported). K3 runs at max reasoning effort by default — no credit card required.

02

Get an API key: Register a developer account at platform.kimi.ai, create an API key, and add credits.

03

Configure OpenAI-compatible client: Set base_url="https://api.moonshot.ai/v1", model ID kimi-k3.

04

OpenRouter: Model ID moonshotai/kimi-k3, Moonshot official pricing with no markup, full 1M context.

05

Enable prompt caching: Set conversation routing cache keys in coding workflows to leverage 90%+ cache hit rates and cut effective input cost to ~$0.55/M.

06

Mark July 27: Full model weights release on Hugging Face; MXFP4/NVFP4 quantized versions and Day-0 vLLM/SGLang support expected. Production deployment needs a 64+ accelerator supernode.

Python
from openai import OpenAI

client = OpenAI(
    api_key="your_moonshot_api_key",
    base_url="https://api.moonshot.ai/v1"
)

response = client.chat.completions.create(
    model="kimi-k3",
    messages=[{"role": "user", "content": "Analyze this code for me..."}]
)

Open-source timeline: WAIC July 17–20 may bring more announcements, then July 27 full K3 weights open-source — the first downloadable open weights above 2 trillion parameters.

05

Which Model Should You Pick? Use Case Matrix and Citable Data

ScenarioRecommended modelWhy
Sustained long coding tasksKimi K3SWE Marathon #1, longest 1M context
Complex repo-level bug fixesClaude Fable 5FrontierSWE 86.6 leads by a wide margin
Terminal/toolchain-heavy AgentsGPT-5.6 SolTerminal Bench 2.1 and Coding Agent Index lead
Ultra-long docs / multimodal understandingKimi K3OmniDocBench 91.1 #1, native vision + 1M context
Cost-sensitive workloadsDeepSeek V4 ProOutput only $3.48/M, far below K3
Self-hosted open weights (post 7/27)Kimi K3Strongest open weights, MXFP4 quantization friendly
A

Parameter scale: 2.8T — nearly 75% above DeepSeek V4 Pro (1.6T), largest downloadable open model (after 7/27).

B

KDA efficiency: KV cache -75%, 1M-token decode +6.3x, training scaling efficiency +2.5x (vs K2).

C

Intelligence Index: Artificial Analysis v4.1 score 57.1 — highest among open models, 2.8 points behind closed-source flagships.

Caveat: Benchmarks above are Moonshot self-reported. K3 uses Kimi Code, GPT uses Codex, Claude uses Claude Code — each with its own harness. Independent third-party replication is still underway; treat as directional guidance.

Kimi K3 is not parameter vanity — KDA, AttnRes, and Stable LatentMoE are real engineering innovations that match or beat some closed-source flagships on long coding tasks and document understanding. API-only setups work for quick starts, but local Macs often hit memory pressure under long Kimi Code Agent loops, and VMs sacrifice Metal/Xcode fidelity. For 7x24 coding Agent uptime, parallel dev servers, and always-on inference, MESHLAUNCH cloud Mac Mini bare-metal rental is usually the better production fit: dedicated Apple Silicon with flexible daily/weekly/monthly terms, ideal for iOS CI/CD and AI Agent automation.

Sources: Moonshot AI official blog · Kimi API Platform docs · Artificial Analysis · OpenRouter pricing · VentureBeat · SCMP

FAQ

Yes. Register a free account at kimi.com to use K3 at max reasoning effort. API calls are paid ($3/$15 per 1M tokens). Budget planning: see our pricing page.

Full weights release on Hugging Face July 27, 2026. Production deployment requires a 64+ accelerator supernode; consumer-grade local deployment is not practical. Model trained with MXFP4 weights + MXFP8 activations — quantization friendly.

K3 has nearly 2x parameters, 8x larger context, and stronger coding and document benchmarks. DeepSeek V4 Pro output costs only $3.48/M — better for cost-sensitive workloads. See the help center for related guidance.

Very useful for analyzing entire codebases in one pass, processing long legal or research documents, and multi-turn Agent long-memory scenarios. Flat pricing with no length surcharge, plus 90%+ cache hit rates, keeps real costs manageable.

Moonshot AI says low and high effort modes will roll out in a future update. Only max mode is available today.