GPUs tracked39Price history3 dayssince Sep 3, 2026Price points (24h)11,753 pointsBiggest 24h dropBiggest 24h riseData updated Sep 5, 2026

How Much VRAM Does Kimi K2 Need to Run Locally?

See devices that run it

Kimi K2 is a MoE model with 1,000B total parameters and 32B active per token. At Q4 it needs 559GB of VRAM to sit entirely on the GPU; with experts offloaded to system RAM it needs only 9GB of VRAM plus 552GB of RAM. No single card in the database fits it; it needs multiple GPUs or the unified memory of a Mac Studio. In offload mode the cheapest card that runs it is the GeForce RTX 2080 Ti, at roughly 4 tokens/s.

Kimi K2 details

Launch dateSep 3, 2025
Revision0905
ArchitectureMoE
GGUF repo
Ollama tag
Downloads (30d)37,308
Min VRAM at Q4559 GB · Q4_K_M · 8K
Experts offloaded9 GB VRAM + 552 GB RAM
Parameters1,000B
MoE32B
Layers61
Hidden size7,168
KV heads64
Head dim192
Max context256K
VendorMoonshot AI

VRAM needed for Kimi K2 by quantization and context

Total at 8K

MoE: all 1,000B parameters must stay resident in memory, while speed is set by the 32B active per token.

QuantizationWeightsGBTotal at 8KGBTotal at 32KGBTotal at 128KGB
Q4_K_M557.4est.558.9560.5567
Q5_K_M674.7est.676.3677.9684.3
Q8_01,036.6est.1,038.11,039.71,046.1
FP161,955.8est.1,957.31,958.91,965.4

Total = weights + KV cache (FP16) + 1GB runtime overhead. Weights marked GGUF are measured file sizes from the repository; the rest are estimated from bytes per parameter. Contexts beyond this model's 256K limit show a dash.

Which GPUs can run Kimi K2?

Q4_K_M · 8K context · sorted by eBay used price

No single card in the database fits Kimi K2: it needs at least 559GB of VRAM at Q4_K_M with an 8K context.

A multi-GPU build needs at least 559GB of VRAM in total.

None of the unified-memory machines in the database fit it either.

GPUs that run Kimi K2 with experts offloaded to RAM

Expert weights live in system RAM (needs at least 552 GB); VRAM holds only attention layers, shared experts and the KV cache, at least 9 GB at Q4. Speed is bound by RAM bandwidth (estimated at 70 GB/s) and is far slower than a full-VRAM setup.

Run Kimi K2 with Ollama, llama.cpp or vLLM

Ollama

The Ollama library has no official tag for it yet.

vLLM

vLLM serves the original-precision weights, which needs far more memory than GGUF and usually more than one GPU.

vllm serve moonshotai/Kimi-K2-Instruct-0905

File names follow that week's model-library snapshot; the definitions are in the methodology. Methodology

FAQ

How much VRAM does Kimi K2 need?

At Q4_K_M with an 8K context Kimi K2 needs about 559GB of VRAM; 672GB leaves comfortable headroom.

What is the cheapest GPU that runs Kimi K2?

No single card in the database fits Kimi K2; it needs multiple GPUs or a Mac Studio's unified memory.

Can you run Kimi K2 with Ollama?

Not yet: there is no official Ollama tag and no public GGUF build for it.