Kimi-K3 needs about 744.5 GB of memory at IQ1_M (a 648.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 9 quants, from 466.4 GB (Q1_0) to 1561.2 GB (Q8_K_XL). Longer context needs more: at 32K tokens the KV cache alone adds about 380.3 GB.
Kimi‑K3: 2.8T‑parameter multimodal model with 1‑million‑token context
Kimi‑K3 is a 2.8T‑parameter multimodal model built on Kimi Delta Attention and Attention Residuals. It supports text, images, and video in a single model and offers a 1‑million‑token context window. The model is released under the Kimi K3 License and is available in multiple GGUF quantizations.
Good at
- Native vision and text understanding
- Long‑horizon coding and tool orchestration
- Agentic knowledge work with interactive visualizations
Limits
- Requires large VRAM for full precision (≈1.6 GB per 8 k context)
- No explicit language support listed
- License is not standard open source
Best for
- Large‑scale research projects
- Complex coding sessions with tool integration
- Multimodal content creation
Who it’s for. Researchers and developers needing a multimodal model with a very long context window. The model is suitable for projects that can accommodate its large memory footprint and licensing terms.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q1_0 | 466.4 GB (11 parts) | 561.9 GB | Too big | Get file → | |
| TQ1_0 | 508.9 GB (12 parts) | 604.5 GB | Too big | Get file → | |
| TQ2_0 | 551.5 GB (13 parts) | 647.0 GB | Too big | Get file → | |
| IQ1_S | 594.0 GB (14 parts) | 689.6 GB | Too big | Get file → | |
| IQ1_M | 648.9 GB (15 parts) | 744.5 GB | Too big | Get file → | |
| IQ2_XXS | 711.1 GB (16 parts) | 806.6 GB | Too big | Get file → | |
| Q2_K_XL | 861.3 GB (19 parts) | 956.9 GB | Too big | Get file → | |
| Q4_K_XL | 1508.7 GB (32 parts) | 1604.2 GB | Too big | Get file → | |
| Q8_K_XL | 1561.2 GB (34 parts) | 1656.7 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_K_XL (file size 1.5 GB, estimated 1.6 GB memory at 8 k context). For 12 GB VRAM, Q8_K_XL (1.56 GB, ~1.66 GB memory) is feasible. Higher‑precision quants require more VRAM and are not recommended for limited budgets.
What GPU do you need for Kimi-K3?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which Kimi-K3 quant should you choose?
IQ1_M (648.9 GB) is the middle option in this repository. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
If memory is tight, the smallest file is Q1_0 at 466.4 GB, with the quality trade-offs that come with fewer bits. Lower bit counts mean smaller files and more quality loss.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Kimi-K3
How much VRAM does Kimi-K3 need?
About 744.5 GB at IQ1_M with an 8K context window. That is the 648.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Kimi-K3 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 561.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Kimi-K3 GGUF?
It depends on the quant: Q1_0 is 466.4 GB, TQ1_0 is 508.9 GB, TQ2_0 is 551.5 GB, IQ1_S is 594.0 GB, IQ1_M is 648.9 GB, IQ2_XXS is 711.1 GB and 3 more.
What context length does Kimi-K3 support?
The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 95.1 GB at 8K and 380.3 GB at 32K.
Where can I download Kimi-K3 GGUF?
The files are on Hugging Face in the unsloth/Kimi-K3-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Kimi-K3 use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Kimi‑K3 on 8 GB VRAM?
Yes, using the Q4_K_XL quantization which requires about 1.6 GB of VRAM at 8 k context.
Does Kimi‑K3 support vision?
Yes, it has native vision support via the MoonViT‑V2 encoder.
What is the context length of Kimi‑K3?
The model supports a 1‑million‑token context window.