Kolibri 1 needs about 48.8 GB of memory at Q4_K_M (a 47.5 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 3 quants, from 47.5 GB (Q4_K_M) to 83.1 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 3.4 GB.
Kolibri 1 – a mixture‑of‑experts LLM for German and English
Kolibri 1 is a 78.1 billion‑parameter MoE model optimized for long‑context reasoning and tool calling. It supports up to 262,144‑token native context and can be extended beyond that. The model is released under Apache‑2.0 and requires a Kolibri‑patched llama.cpp to run.
Good at
- Long‑context reasoning
- Tool calling
- German‑English bilingual tasks
Limits
- Requires Kolibri‑patched llama.cpp
- Full model must stay in memory
- No vision support
Best for
- Document processing
- Multi‑step reasoning
- Agentic tool workflows
Who it’s for. Developers building conversational assistants or internal drafting systems that need robust German and English reasoning. The model is suitable for environments where outputs are reviewed before action.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 47.5 GB | 48.8 GB | Too big | Get file → | |
| Q6_K | 64.2 GB (2 parts) | 65.6 GB | Too big | Get file → | |
| Q8_0 | 83.1 GB (2 parts) | 84.5 GB | Too big | Get file → |
For 48.8 GB memory at 8k context, use Q4_K_M (47.45 GB file). Q6_K (64.23 GB) and Q8_0 (83.14 GB) require larger memory. Shards are available for Q6_K and Q8_0. 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: 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 Kolibri 1 quant should you choose?
Q4_K_M is the usual starting point for Kolibri 1: it balances file size and quality well. Here it is a 47.5 GB file.
The Q8_0 file is 83.1 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.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Kolibri 1
How much VRAM does Kolibri 1 need?
About 48.8 GB at Q4_K_M with an 8K context window. That is the 47.5 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 48.8 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: Q4_K_M is 47.5 GB, Q6_K is 64.2 GB, Q8_0 is 83.1 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 Prompt48/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?
No. The smallest quant requires 48.8 GB memory at 8k context, far above 8 GB.
Does Kolibri 1 support vision tasks?
No. The model is text‑only and has no vision capability.
What license applies to Kolibri 1 GGUF files?
The GGUF files are released under Apache‑2.0, matching the original model license.