Moonshot AI/Mixture of Experts

MKimi K2.7 Code

Name: Kimi K2.7 Code
Author: Moonshot AI

Kimi K2.7 Code — the coding-specialised K2.7: 1T total, 32B active, MLA attention, 256K context.

chatcodingreasoningagentictool_usemultilingual

1026.88B

Parameters (32B active)

256K

Context length

Benchmarks

Quantizations

568K

HF downloads

Architecture

MoE

Released

2026-06-12

Layers

KV Heads

Head Dim

112

Family

kimi

Quantization Options

Context length:

Quant	Bits	VRAM @ 16K	Quality
IQ2_XXS	2.38	306.8 GB 306.0 + 0.8 KV	low
IQ2_M	2.93	377.4 GB 376.6 + 0.8 KV	low
Q2_K	3.16	406.9 GB 406.1 + 0.8 KV	low
IQ3_XXS	3.25	418.5 GB 417.7 + 0.8 KV	low
IQ3_XS	3.5	450.6 GB 449.7 + 0.8 KV	low
Q3_K_S	3.64	468.5 GB 467.7 + 0.8 KV	low
IQ3_M	3.76	483.9 GB 483.1 + 0.8 KV	low
Q3_K_M	4	514.7 GB 513.9 + 0.8 KV	low
Q3_K_L	4.3	553.2 GB 552.4 + 0.8 KV	moderate
IQ4_XS	4.46	573.8 GB 573.0 + 0.8 KV	moderate
Q4_K_S	4.67	600.7 GB 599.9 + 0.8 KV	moderate
Q4_K_M	4.89	629.0 GB 628.2 + 0.8 KV	good
Q5_K_S	5.57	716.3 GB 715.5 + 0.8 KV	good
Q5_K_M	5.7	732.9 GB 732.1 + 0.8 KV	good
Q6_K	6.56	843.3 GB 842.5 + 0.8 KV	excellent
Q8_0	8.5	1092.4 GB 1091.5 + 0.8 KV	lossless
FP16	16	2055.1 GB 2054.2 + 0.8 KV	lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Too big for a single GPU — plan a multi-GPU deployment

Even the lightest quant needs ~307 GB. Size GPUs, replicas, TCO and scaling for a production setup. Open in Enterprise →

▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN KIMI K2.7 CODE NOW

Rent on Vast.ai →Or RunPod →

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (9)

τ²-Bench90.1

GPQA Diamond89.6

AA Long Context75.0

IFBench63.1

AA Coding60.8

SciCode47.5

Terminal-Bench44.7

AA Intelligence43.0

HLE35.0

Run this model

▸Easiest way to get started·Beginners

DOCS ↗

curl -fsSL https://ollama.com/install.sh | sh

$ollama run kimi:1027b-q4_K_M

Tag may need adjustment — check ollama.com/library/kimi for available tags.

▸ SETUP GUIDE

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more

Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

HuggingFace GGUF Downloads Build Hardware

Find the best GPU for Kimi K2.7 Code

Build Hardware for Kimi K2.7 Code

QUANT	BPW	VRAM	QUALITY
IQ2_XXS	2.38	306.0 GB	65%
IQ2_M	2.93	376.6 GB	75%
Q2_K	3.16	406.1 GB	78%
IQ3_XXS	3.25	417.7 GB	82%
IQ3_XS	3.5	449.7 GB	84%
Q3_K_S	3.64	467.7 GB	85%
IQ3_M	3.76	483.1 GB	86%
Q3_K_M	4	513.9 GB	88%
Q3_K_L	4.3	552.4 GB	90%
IQ4_XS	4.46	573.0 GB	92%
Q4_K_S	4.67	599.9 GB	93%
Q4_K_M	4.89	628.2 GB	94%
Q5_K_S	5.57	715.5 GB	96%
Q5_K_M	5.7	732.1 GB	96%
Q6_K	6.56	842.5 GB	97%
Q8_0	8.5	1091.5 GB	100%
FP16	16	2054.2 GB	100%