KI

Kimi-K2-Thinking

KimiGGUF by unslothUpdated 2026-01-27
Download on Hugging Face
Parameters1030B
Context256K
Layers61
Architecturedeepseek2
Smallest quant246.8 GB
Quick answer

Kimi-K2-Thinking needs about 622.8 GB of memory at Q4_K_M (a 621.2 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 27 quants, from 246.8 GB (TQ1_0) to 2053.2 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

Kimi‑K2‑Thinking: 1 T parameter deep‑seek model with advanced reasoning and tool orchestration

Kimi‑K2‑Thinking is a 1 T‑parameter deep‑seek model trained for step‑by‑step reasoning and dynamic tool use. It supports a 262 144‑token context window and is available in multiple GGUF quantizations. The model is part of the Kimi family and was quantized by unsloth.

Good at

  • Multi‑step reasoning
  • Tool orchestration
  • Long‑horizon agentic tasks

Limits

  • Large memory requirement for high‑precision quants
  • No vision capability
  • License is other

Best for

  • Research workflows requiring many tool calls
  • Complex coding tasks with step‑by‑step logic
  • Long‑form reasoning and analysis

Who it’s for. Developers and researchers who need a large, reasoning‑focused model that can invoke tools over many steps. The model is suitable for local deployment with sufficient GPU memory.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
TQ1_0246.8 GB (6 parts) 248.4 GB
Too big Get file →
IQ1_S285.3 GB (6 parts) 286.9 GB
Too big Get file →
IQ1_M309.3 GB (7 parts) 310.9 GB
Too big Get file →
IQ2_XXS335.2 GB (7 parts) 336.8 GB
Too big Get file →
IQ2_M353.6 GB (8 parts) 355.1 GB
Too big Get file →
Q2_K374.1 GB (8 parts) 375.7 GB
Too big Get file →
Q2_K_L374.4 GB (8 parts) 376.0 GB
Too big Get file →
Q2_K_XL386.3 GB (8 parts) 387.9 GB
Too big Get file →
IQ3_XXS421.7 GB (9 parts) 423.3 GB
Too big Get file →
Q3_K_S443.2 GB (10 parts) 444.7 GB
Too big Get file →
Q3_K_XL455.1 GB (10 parts) 456.7 GB
Too big Get file →
Q3_K_M490.0 GB (11 parts) 491.6 GB
Too big Get file →
IQ4_XS547.2 GB (12 parts) 548.8 GB
Too big Get file →
IQ4_NL579.1 GB (12 parts) 580.7 GB
Too big Get file →
Q4_0581.2 GB (13 parts) 582.8 GB
Too big Get file →
Q4_K_S583.3 GB (13 parts) 584.9 GB
Too big Get file →
Q4_K_M621.2 GB (13 parts) 622.8 GB
Too big Get file →
Q4_1642.9 GB (13 parts) 644.5 GB
Too big Get file →
Q4_K_XL646.2 GB (14 parts) 647.8 GB
Too big Get file →
Q5_K_S707.0 GB (15 parts) 708.6 GB
Too big Get file →
Q5_K_M728.7 GB (16 parts) 730.3 GB
Too big Get file →
Q5_K_XL729.8 GB (16 parts) 731.4 GB
Too big Get file →
Q6_K842.9 GB (18 parts) 844.5 GB
Too big Get file →
Q6_K_XL879.1 GB (19 parts) 880.6 GB
Too big Get file →
Q8_01091.1 GB (23 parts) 1092.6 GB
Too big Get file →
Q8_K_XL1190.1 GB (25 parts) 1191.6 GB
Too big Get file →
BF162053.2 GB (46 parts) 2054.7 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_S (≈584 GB memory at 8k context). For 16 GB VRAM, Q4_K_M (≈623 GB). For 24 GB VRAM, Q5_K_S (≈709 GB). Larger quants such as Q6_K or BF16 exceed typical consumer GPUs.

What GPU do you need for Kimi-K2-Thinking?

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-K2-Thinking quant should you choose?

Q4_K_M is the usual starting point for Kimi-K2-Thinking: it balances file size and quality well. Here it is a 621.2 GB file.

The Q8_0 file is 1091.1 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is TQ1_0 at 246.8 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

Architecturedeepseek2
Parameters1030B (approximate, mixture of experts)
Layers61
Trained context262,144 tokens
KV cache (fp16)130 KB per token

Source

Quantized byunsloth
FamilyKimi
Licenseother
Downloads (HF)4,449
Last updated2026-01-27

Frequently asked questions about Kimi-K2-Thinking

How much VRAM does Kimi-K2-Thinking need?

About 622.8 GB at Q4_K_M with an 8K context window. That is the 621.2 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-K2-Thinking on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 248.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Kimi-K2-Thinking GGUF?

It depends on the quant: TQ1_0 is 246.8 GB, IQ1_S is 285.3 GB, IQ1_M is 309.3 GB, IQ2_XXS is 335.2 GB, IQ2_M is 353.6 GB, Q2_K is 374.1 GB and 21 more.

What context length does Kimi-K2-Thinking support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.

Where can I download Kimi-K2-Thinking GGUF?

The files are on Hugging Face in the unsloth/Kimi-K2-Thinking-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-K2-Thinking 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‑K2‑Thinking on 8 GB VRAM?

Use the Q4_K_S quant, which requires about 584 GB of VRAM at 8k context. This is the smallest quant that fits 8 GB VRAM.

Does the model support image input?

No, Kimi‑K2‑Thinking does not have vision capabilities.

What is the maximum context length?

The model supports up to 262,144 tokens of context.

Other GGUFs of Kimi-K2-Thinking

Similar models

Browse all →

More from unsloth

All by unsloth →

More Kimi models

All Kimi →

Other large models

All large models →