Language

Kimi K2.6 (Hybrid Thinking, MoE)

Moonshot AI Kimi K2.6 hybrid-thinking MoE — 1T total params, 256K context. Replica-routed on B200-class infrastructure; Unsloth measures >40 t/s on B200. Recommended `UD-Q2_K_XL` (350GB) for size/quality balance.

MITSource verified

Specification

Model ID

kimi-k2.6

Family

kimi

Modality

Language

Parameters

1T (MoE, 32B active)

Context

262,144 tokens

Quantization

UD-Q4_K_XL

Weights

558.8 GB

Minimum RAM

400 GB

Source

Weights and provenance.

Repository

unsloth/Kimi-K2.6-GGUF

Registry license

MIT

Source license

other

Access

Open — weights fetch without accepting additional terms.

The registry records this model as MIT 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-k2.6

# Serve it
tenzro model serve kimi-k2.6

# 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-k2.6"}]}'

Same family

Kimi K2 Instruct (MoE)

1T (MoE, 32B active, code-focused)

← All models