Language

Kimi K3 (MoE, multimodal)

Moonshot AI Kimi K3 — 2.8T total parameters, 104B active, 896 routed experts with 16 selected per token and 2 shared. Kimi Delta Attention plus gated MLA across 93 layers, 1M context, 160K vocabulary, MXFP4 weights and MXFP8 activations from quantization-aware training. Text, image, and video via the MoonViT-V2 encoder. `UD-IQ1_S` (594GB) is the smallest quant; `UD-Q2_K_XL` (861GB) is the size/quality balance point. Both exceed any single machine, so whole-model serving means a pipeline cluster; a lone host runs it as distributed expert extraction instead.

Kimi K3 LicenseSource verified

Specification

Model ID

kimi-k3

Family

kimi-k3

Modality

Language

Parameters

2.8T total / 104B active (MoE)

Context

1,048,576 tokens

Quantization

UD-IQ1_S

Weights

553.2 GB

Minimum RAM

610 GB

Source

Weights and provenance.

Repository

unsloth/Kimi-K3-GGUF

Registry license

Kimi K3 License

Source license

other

Access

Open — weights fetch without accepting additional terms.

The registry records this model as Kimi K3 License while the source repository states other. Both are shown here rather than one being preferred. Treat the stricter of the two as binding until the discrepancy is resolved upstream.

Serve it

# Pull the weights onto a node
tenzro model download kimi-k3

# Serve it
tenzro model serve kimi-k3

# Or query the registry entry over JSON-RPC
curl https://rpc.tenzro.xyz \
  -X POST -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,
       "method":"tenzro_getModel",
       "params":[{"model_id":"kimi-k3"}]}'

← All models