KO

Kolibri 1

KolibriGGUF by webmp3Updated 2026-10-08
Download on Hugging Face
Parameters78B
Context256K
Layers50
Architecturekolibri1
Smallest quant22.5 GB
Quick answer

Kolibri 1 needs about 32.2 GB of memory at IQ3_XXS (a 30.9 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 3 quants, from 22.5 GB (IQ2_XS) to 40.9 GB (IQ4_XS). Longer context needs more: at 32K tokens the KV cache alone adds about 3.4 GB.

Kolibri 1 – 78B Mixture‑of‑Experts model for German and English

Kolibri 1 is a 78B parameter mixture‑of‑experts model optimized for long‑context inference. It supports explicit reasoning and tool calling in German and English. The model is available in three GGUF quantizations.

Good at

  • Long‑context reasoning
  • Tool‑calling
  • Multi‑step coding

Limits

  • Requires full model in memory
  • No vision support
  • Needs patched llama.cpp

Best for

  • Document‑processing systems
  • Conversational assistants
  • Agentic workflows

Who it’s for. Developers building conversational assistants or document‑processing pipelines that need German and English reasoning. The model is suitable for systems where a human reviews outputs before action.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ2_XS22.5 GB 23.8 GB
Too big Get file →
IQ3_XXS30.9 GB 32.2 GB
Too big Get file →
IQ4_XS40.9 GB 42.2 GB
Too big Get file →

For 8 GB VRAM use IQ2_XS (22.49 GB file, ~23.8 GB memory at 8 k context). For 12 GB VRAM use IQ3_XXS (30.87 GB file, ~32.2 GB memory). For 16 GB VRAM use IQ4_XS (40.87 GB file, ~42.2 GB memory). Memory estimates are approximate.

What GPU do you need for Kolibri 1?

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: the best quant that fits is IQ2_XS (22.5 GB file, about 23.8 GB in use).
  • 32 GB VRAM: the best quant that fits is IQ2_XS (22.5 GB file, about 23.8 GB in use).
  • 48 GB VRAM: the best quant that fits is IQ4_XS (40.9 GB file, about 42.2 GB in use).

Which Kolibri 1 quant should you choose?

IQ3_XXS (30.9 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

If memory is tight, the smallest file is IQ2_XS at 22.5 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

Architecturekolibri1
Parameters78B (approximate, mixture of experts)
Layers50
Trained context262,144 tokens
KV cache (fp16)100 KB per token

Source

Quantized bywebmp3
FamilyKolibri
Licenseapache-2.0
Downloads (HF)2,472
Last updated2026-10-08

Frequently asked questions about Kolibri 1

How much VRAM does Kolibri 1 need?

About 32.2 GB at IQ3_XXS with an 8K context window. That is the 30.9 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 Kolibri 1 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 23.8 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Kolibri 1 GGUF?

It depends on the quant: IQ2_XS is 22.5 GB, IQ3_XXS is 30.9 GB, IQ4_XS is 40.9 GB.

What context length does Kolibri 1 support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.8 GB at 8K and 3.4 GB at 32K.

Where can I download Kolibri 1 GGUF?

The files are on Hugging Face in the webmp3/Sakura-MicroQuality-Kolibri-1-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 Kolibri 1 use?

Hugging Face lists the license as apache-2.0. 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 Kolibri 1 on 8 GB VRAM?

Use the IQ2_XS quantization; it requires about 23.8 GB memory at 8 k context, so 8 GB VRAM is insufficient.

Does Kolibri 1 support vision tasks?

No, the model is text‑only and does not include vision capabilities.

What license applies to Kolibri 1?

The model is released under the apache-2.0 license.

Other GGUFs of Kolibri-1

Similar models

Browse all →

More from webmp3

All by webmp3 →

More Kolibri models

All Kolibri →

Other large models

All large models →