Kimi · Over 80B · Mixture of experts
Kimi K2 Instruct
1026.41 billion parameters in total, but only 39.06 billion are read for each token. That split is the whole point of the design: it costs the memory of a large model and the speed of a small one. It compresses its cache into a latent vector, which keeps the cache small at long context.
About Kimi K2 Instruct
Kimi K2 Instruct is a mixture-of-experts model with 1026.41B parameters, 39.06B of them active for each token, built from 61 layers with a hidden size of 7168. Its repository was published on 2025-07-11. The license listed on Hugging Face is other. It compresses its cache into a latent vector, which keeps the cache small at long context.
Memory needed, by quantization
At an 8k context window. Quality is our rough ranking of how much the compression costs you: anything at or above Q5 is hard to tell apart from the original in normal use.
| Quantization | File size | KV cache | Total VRAM | Quality | What it costs you |
|---|---|---|---|---|---|
| Q8_0 | 1015.66 GB | 0.54 GB | 1015.55 GB | 99% | Lossless in practice. Use it when the memory is there. |
| Q6_K | 783.85 GB | 0.54 GB | 784 GB | 98% | Very close to Q8 for two thirds of the size. |
| Q5_K_M | 679.9 GB | 0.54 GB | 680.16 GB | 96% | The quality-first choice when Q6 will not fit. |
| Q4_K_M | 577.14 GB | 0.54 GB | 577.52 GB | 93% | The default. Best size-to-quality ratio for local use. |
| IQ4_XS | 507.83 GB | 0.54 GB | 508.29 GB | 91% | Importance-matrix 4-bit. Q4_K_S quality, smaller file. |
| Q3_K_M | 467.21 GB | 0.54 GB | 467.71 GB | 86% | Degradation starts to show. A way to fit one size up. |
| IQ3_XXS | 365.64 GB | 0.54 GB | 366.26 GB | 79% | Aggressive. Only worth it on very large models. |
What a longer conversation costs
Same model at Q4_K_M, only the context window changes. This model compresses its cache into a latent vector, which is why the numbers stay small at long context.
| Context | KV cache | Total VRAM | Fits in 8 GB | Fits in 12 GB | Fits in 24 GB |
|---|---|---|---|---|---|
| 4k | 0.27 GB | 577.25 GB | no | no | no |
| 8k | 0.54 GB | 577.52 GB | no | no | no |
| 16k | 1.07 GB | 578.06 GB | no | no | no |
| 32k | 2.14 GB | 579.13 GB | no | no | no |
| 64k | 4.29 GB | 581.27 GB | no | no | no |
| 128k | 8.58 GB | 585.56 GB | no | no | no |
The "fits" columns allow for the roughly 0.8 GB Windows keeps for the desktop.
Which hardware runs Kimi K2 Instruct
0 of 118 consumer devices run it at a quantization worth using, at an 8k context window.
| Device | Memory | Verdict | Quantization | Used | Speed |
|---|---|---|---|---|---|
| Apple M3 Ultra 512GB Apple | 512 GB | Just fits | n/a | 468.22 GB | 17.7 t/s |
| RTX PRO 6000 Blackwell NVIDIA | 96 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.8 t/s |
| GeForce RTX 5090 NVIDIA | 32 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.5 t/s |
| RTX 6000 Ada Generation NVIDIA | 48 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.5 t/s |
| RTX A6000 NVIDIA | 48 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.5 t/s |
| Radeon PRO W7900 AMD | 48 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.5 t/s |
| GeForce RTX 5080 NVIDIA | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 5070 Ti NVIDIA | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 4090 NVIDIA | 24 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 4080 SUPER NVIDIA | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 4080 NVIDIA | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 3090 Ti NVIDIA | 24 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 3090 NVIDIA | 24 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 3080 Ti NVIDIA | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 3080 12GB NVIDIA | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 5090 Laptop NVIDIA | 24 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 5080 Laptop NVIDIA | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| RTX A5000 NVIDIA | 24 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| Radeon RX 7900 XTX AMD | 24 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| Radeon RX 7900 XT AMD | 20 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.4 t/s |
| GeForce RTX 5070 NVIDIA | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.3 t/s |
| GeForce RTX 4070 Ti SUPER NVIDIA | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.3 t/s |
| GeForce RTX 4070 Ti NVIDIA | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.3 t/s |
| GeForce RTX 4070 SUPER NVIDIA | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.3 t/s |
| GeForce RTX 4070 NVIDIA | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.3 t/s |
Show 35 more devices
| GeForce RTX 4090 Laptop NVIDIA | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.3 t/s |
| Radeon RX 9070 XT AMD | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.3 t/s |
| Radeon RX 9070 AMD | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.3 t/s |
| Radeon RX 7900 GRE AMD | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.3 t/s |
| Radeon RX 7800 XT AMD | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.3 t/s |
| GeForce RTX 5060 Ti 16GB NVIDIA | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.2 t/s |
| GeForce RTX 3060 12GB NVIDIA | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.2 t/s |
| GeForce RTX 2060 12GB NVIDIA | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.2 t/s |
| GeForce RTX 4080 Laptop NVIDIA | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.2 t/s |
| RTX A4000 NVIDIA | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.2 t/s |
| Radeon RX 7700 XT AMD | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.2 t/s |
| Radeon RX 6900 XT AMD | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.2 t/s |
| Radeon RX 6800 AMD | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.2 t/s |
| Arc B580 Intel | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.2 t/s |
| Arc A770 16GB Intel | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.2 t/s |
| GeForce RTX 4060 Ti 16GB NVIDIA | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.1 t/s |
| Radeon RX 6700 XT AMD | 12 GB | Partial offload 61/61 | n/a | 577.52 GB | 2.1 t/s |
| Radeon RX 7600 XT AMD | 16 GB | Partial offload 61/61 | n/a | 577.52 GB | 2 t/s |
| GeForce RTX 2080 Ti NVIDIA | 11 GB | Partial offload 1/61 | n/a | 577.52 GB | 1.4 t/s |
| GeForce GTX 1080 Ti NVIDIA | 11 GB | Partial offload 1/61 | n/a | 577.52 GB | 1.4 t/s |
| GeForce RTX 5060 Ti 8GB NVIDIA | 8 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 5060 NVIDIA | 8 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 4060 Ti 8GB NVIDIA | 8 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 4060 NVIDIA | 8 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 3080 10GB NVIDIA | 10 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 3070 Ti NVIDIA | 8 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 3070 NVIDIA | 8 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 3060 Ti NVIDIA | 8 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 3050 8GB NVIDIA | 8 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 3050 6GB NVIDIA | 6 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce GTX 1660 SUPER NVIDIA | 6 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce GTX 1650 NVIDIA | 4 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 4070 Laptop NVIDIA | 8 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| GeForce RTX 4060 Laptop NVIDIA | 8 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
| Arc B570 Intel | 10 GB | Will not run | n/a | 577.52 GB | 1.3 t/s |
How it compares
Other models in the same family or of a similar size, on a 12 GB card and a 24 GB card, each at the best quantization that fits.
| Model | Weights at 4 bits | GeForce RTX 3060 12GB | GeForce RTX 4090 |
|---|---|---|---|
| Kimi K2 Instruct | 577.1 GB | does not fit well | does not fit well |
| Kimi K2.6 | 577.4 GB | does not fit well | does not fit well |
| DeepSeek R1 | 384.9 GB | does not fit well | does not fit well |
| DeepSeek V3.2 | 385.4 GB | does not fit well | does not fit well |
| GLM 5.2 | 423.6 GB | does not fit well | does not fit well |
What to buy to run Kimi K2 Instruct
Nothing we track runs this model comfortably at 8k context, so there is no honest buying answer here.
At 8k context this model needs more memory than anything in our consumer catalog can give it at a quantization we would recommend. The realistic options are a shorter context window, a smaller model from the same family below, or multiple cards, which we do not model.
Where it sits in the catalog
Source: moonshotai/Kimi-K2-Instruct on Hugging Face. Downloaded 166k times in the last month. Published 2025-07-11. Architecture figures are read from the repository's own configuration file, so they move when the model does.
Questions people ask about Kimi K2 Instruct
How much VRAM does Kimi K2 Instruct need?
At Q4_K_M with an 8k context window, Kimi K2 Instruct needs about 577.52 GB: 576.48 GB of weights, 0.54 GB of KV cache and 0.51 GB of overhead. At the highest-quality file it needs about 1015.55 GB.
What is the cheapest GPU that runs Kimi K2 Instruct?
No consumer device in our catalog fits this model entirely.
How much does a longer context window add for Kimi K2 Instruct?
On the memory side, a 32k window adds 2.1 GB of KV cache on top of the weights, and the cache grows more slowly than a conventional model because of its attention design. The longest window the model supports natively is 128k tokens.