KI

Kimi Linear 48B A3B Instruct

KimiGGUF by ymckiUpdated 2026-01-27
Download on Hugging Face
Parameters49.1B
Context1024K
Layers27
Architecturekimi-linear
Smallest quant18.0 GB
Quick answer

Kimi Linear 48B A3B Instruct needs about 45.6 GB of memory at Q4_K_M (a 29.7 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 6 quants, from 18.0 GB (Q2_K) to 34.9 GB (Q5_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 61.6 GB.

Kimi Linear 48B A3B Instruct – a large hybrid linear attention model

Kimi Linear 48B A3B Instruct is a 49.1 billion‑parameter model with 3 billion active parameters. It uses a hybrid KDA‑to‑MLA architecture and supports up to 1 048 576 context tokens. The model is released under the MIT license.

Good at

  • Long‑context inference
  • High decoding throughput
  • Tool‑use capability

Limits

  • Requires large VRAM for full KV cache
  • No vision support
  • Large memory footprint for high‑quality quants

Best for

  • Long‑form text generation
  • Research on linear attention
  • Fine‑tuning with limited GPU memory

Who it’s for. Researchers and developers needing a large, efficient linear attention model for long‑context tasks. The model is suitable for those who can allocate sufficient GPU and CPU resources to handle its memory demands.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K18.0 GB 33.9 GB
Too big Get file →
IQ3_M21.5 GB 37.4 GB
Too big Get file →
IQ4_XS26.3 GB 42.2 GB
Too big Get file →
MXFP4_MOE27.2 GB 43.1 GB
Too big Get file →
Q4_K_M29.7 GB 45.6 GB
Too big Get file →
Q5_K_M34.9 GB 50.8 GB
Too big Get file →

For 8 GB VRAM, use Q2_K (18.03 GB file, ~33.9 GB memory at 8 k context). For 16 GB VRAM, Q4_K_M (29.7 GB file, ~45.6 GB memory) is a good balance. For 24 GB VRAM, Q5_K_M (34.87 GB file, ~50.8 GB memory) offers higher quality.

What GPU do you need for Kimi Linear 48B A3B Instruct?

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: the best quant that fits is Q4_K_M (29.7 GB file, about 45.6 GB in use).

Which Kimi Linear 48B A3B Instruct quant should you choose?

Q4_K_M is the usual starting point for Kimi Linear 48B A3B Instruct: it balances file size and quality well. Here it is a 29.7 GB file.

If memory is tight, the smallest file is Q2_K at 18.0 GB, with the quality trade-offs that come with fewer bits. 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.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturekimi-linear
Parameters49.1B (approximate, mixture of experts)
Active parameters3B
Layers27
Trained context1,048,576 tokens
KV cache (fp16)1836 KB per token

Source

Quantized byymcki
FamilyKimi
Licensemit
Downloads (HF)666
Last updated2026-01-27

Frequently asked questions about Kimi Linear 48B A3B Instruct

How much VRAM does Kimi Linear 48B A3B Instruct need?

About 45.6 GB at Q4_K_M with an 8K context window. That is the 29.7 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 Linear 48B A3B Instruct on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 33.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Kimi Linear 48B A3B Instruct GGUF?

It depends on the quant: Q2_K is 18.0 GB, IQ3_M is 21.5 GB, IQ4_XS is 26.3 GB, MXFP4_MOE is 27.2 GB, Q4_K_M is 29.7 GB, Q5_K_M is 34.9 GB.

What context length does Kimi Linear 48B A3B Instruct 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 15.4 GB at 8K and 61.6 GB at 32K.

Where can I download Kimi Linear 48B A3B Instruct GGUF?

The files are on Hugging Face in the ymcki/Kimi-Linear-48B-A3B-Instruct-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 Linear 48B A3B Instruct use?

Hugging Face lists the license as mit. 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 Linear 48B A3B Instruct on 8 GB VRAM?

Only the Q2_K quant can be loaded; it requires ~33.9 GB memory at 8 k context, so it will need CPU fallback for KV cache.

Does the model support vision tasks?

No, the model does not include vision capabilities.

What license applies to this model?

The model is released under the MIT license.

Other GGUFs of Kimi-Linear-48B-A3B-Instruct

Similar models

Browse all →

More from ymcki

All by ymcki →

More Kimi models

All Kimi →

Other large models

All large models →