AI agents · OpenClaw · self-hosting · automation

Quick Answer

Kimi K3 Open Weights: What It Costs to Self-Host

Published:

The Short Answer

Kimi K3’s full weights release by July 27, 2026 under a Modified MIT license — but “open weights” does not mean “runs on your laptop.” At 2.8 trillion parameters (Mixture-of-Experts), Kimi K3 is the first open model in the 3T-parameter class, and self-hosting it is a data-center project, not a weekend one.

The model has been available on the Kimi API since July 16, 2026 at $3 input / $15 output per MTok (flat pricing, no peak/valley surge). The July 27 weights drop is about control — inspection, fine-tuning, air-gapped deployment — not about saving money.

What Kimi K3 Actually Is (July 2026)

SpecKimi K3
DeveloperMoonshot AI
Total parameters2.8 trillion (MoE)
Context window1M tokens
ArchitectureKimi Delta Attention (KDA) + Attention Residuals
ModalityText + native visual understanding
API pricing$3 / $15 per MTok (flat)
API launchJuly 16, 2026
Weights releaseBy July 27, 2026
LicenseModified MIT
Artificial Analysis Index~57

Kimi K3 leads the Arena.ai Frontend Code Arena ahead of Claude Fable 5 and sits near the top of the open-weight tier on the Artificial Analysis Intelligence Index (~57).

The Hardware Math

The single most important spec for self-hosting is VRAM — the active experts plus KV cache must fit in GPU memory for usable speed.

TierConfigWhat runs
Frontier (native)16x H100 80GB (~1,280GB) or multi-B200Kimi K3 at full quality
Quantized8x H200 / multi-node H100, INT4/MXFP4Kimi K3 with quality loss
Workstation1x RTX 4090 (24GB)NOT Kimi K3 — use GLM-5.2 or DeepSeek V4 Flash

Reality check: even at INT4/MXFP4 quantization, a 2.8T MoE model’s weight footprint is enormous. If your goal is “open weights I can run on a single 24GB card,” Kimi K3 is the wrong model. GLM-5.2 (744B, MIT) or DeepSeek V4 Flash are the practical self-host picks in July 2026.

Self-Host vs API: The Cost Decision

Use the API ($3/$15 per MTok) when:

  • Your volume is low or spiky
  • You can’t keep a 16x-H100 cluster near full utilization
  • You want zero infra maintenance

Self-host the weights when:

  • You need air-gapped / data-residency compliance
  • You’re fine-tuning or modifying the model
  • You have sustained, high-volume, predictable throughput to amortize the cluster

At low-to-medium volume, idle GPU time makes self-hosting more expensive than the API — the classic open-weight trap. The weights give you control; they rarely give you cheaper inference below cluster-saturating scale.

What to Do This Week

  1. If you just want to use K3: the API at $3/$15 is live now — no need to wait for weights.
  2. If you want open weights to run yourself: benchmark GLM-5.2 first; it’s self-hostable today, MIT-licensed, and ~1/3 the output cost of K3.
  3. If you specifically need K3 self-hosted: budget for a multi-node H100/H200 cluster and plan for INT4/MXFP4 quantization from day one.
  4. Watch the license text — “Modified MIT” may carry deployment or attribution conditions; read it before commercial use.

Sources