Can I run Kimi K2 Base locally?
Kimi K2 Base by moonshotai has 1.03T parameters (33B active per token, 384 experts, 8 used per token). At 4-bit with an 8K context it needs about 586 GB of GPU memory. It fits on 1 of 65 GPUs and Macs we track.
Memory needed by quant and context length
GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use parameter count × bits per weight. How we calculate.
| Quant | Quality | 4K ctx | 8K ctx | 32K ctx | 128K ctx |
|---|---|---|---|---|---|
| 2-bit (Q2_K) | rough | 359 GB | 360 GB | 361 GB | 368 GB |
| 3-bit (Q3_K_M) | usable | 468 GB | 468 GB | 470 GB | 476 GB |
| 4-bit (Q4_K_M) | good, most popular | 585 GB | 586 GB | 587 GB | 594 GB |
| 5-bit (Q5_K_M) | very good | 682 GB | 682 GB | 684 GB | 690 GB |
| 6-bit (Q6_K) | near lossless | 785 GB | 785 GB | 787 GB | 793 GB |
| 8-bit (Q8_0) | effectively lossless | 1017 GB | 1017 GB | 1019 GB | 1025 GB |
| 16-bit (F16/BF16) | original weights | 1913 GB | 1913 GB | 1915 GB | 1921 GB |
Which GPUs and Macs run Kimi K2 Base?
Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).
Questions
How much VRAM does Kimi K2 Base need?
About 586 GB at 4-bit (Q4_K_M) with an 8K context: 584 GB of weights (estimated), 0.5 GB of KV cache and 0.7 GB of runtime overhead. At 8-bit it needs about 1017 GB.
What is the smallest GPU that runs Kimi K2 Base?
No single consumer GPU in our table fits it at Q4_K_M with 8K context.
Can I run Kimi K2 Base on a Mac?
Not at Q4_K_M with 8K context on any Mac in our table, even a 512 GB Mac Studio, using the default GPU memory limit.
Similar models
Same family or a similar parameter count, with the VRAM each needs at Q4_K_M and 8K context.
Kimi K2 Instruct
1.03T · 579 GB at 4-bitKimi K2 Thinking
1.03T · 582 GB at 4-bitKimi K2 Instruct 0905
1.03T · 582 GB at 4-bitMiMo V2.6 Pro RL
1.02T · 587 GB at 4-bitMiMo V2.5 Pro
1.02T · 584 GB at 4-bitLing 1T
1000B · 570 GB at 4-bitEmbed this
Free badge and "Will it run on my GPU?" widget for model cards, READMEs and blogs. No cookies, no tracking. All options.
Copy a snippet
Markdown badge
[](https://runsonmy.com/model/kimi-k2-base/) Widget (script)
<div class="runsonmy-widget" data-model="kimi-k2-base"><a href="https://runsonmy.com/model/kimi-k2-base/">Will Kimi K2 Base run on my GPU? Check on RunsOnMy</a></div>
<script async src="https://runsonmy.com/embed.js"></script> Widget (iframe)
<iframe src="https://runsonmy.com/embed/m/kimi-k2-base/" width="100%" height="220" style="max-width:440px;border:0" loading="lazy" title="Will Kimi K2 Base run on my GPU? (RunsOnMy)"></iframe> Model facts
| Hugging Face repo | moonshotai/Kimi-K2-Base |
| Parameters | 1,026,408,235,864 (1.03T) · ~33B active (computed from config) |
| Layers / hidden size | 61 / 7168 |
| Attention | MLA; 64 heads, 64 KV heads, head dim 112 |
| KV cache per token (FP16) | 68.6 KiB |
| Max context | 131,072 tokens |
| License | other |
| Released on HF | Jul 3, 2025 |
| Downloads (30d) / likes | 14,155 / 306 |
Sources: Hugging Face model API · config.json (moonshotai/Kimi-K2-Base) · GPU specs from vendor pages (list). Data fetched Nov 18, 2026. Spot an error? Report it.