HI

Hikari Bonsai 27B Nimble LoRA

HikariGGUF by zapaboboujUpdated 2026-10-04
Download on Hugging Face
Parameters26.9B
Context256K
Layers64
Architectureqwen35
Smallest quant7.2 GB
Quick answer

Hikari Bonsai 27B Nimble LoRA needs about 9.9 GB of memory at its main file (a 7.2 GB file) with an 8K context window, so it fits on a 12 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.

Hikari Bonsai 27B Nimble LoRA – 27B Qwen35 model adapted with LoRA

Hikari Bonsai 27B Nimble LoRA is a 26.9B parameter Qwen35 model adapted through LoRA training. The GGUF artefact is 7.21 GB and requires about 9.9 GB of memory at 8 k context. It is released under apache‑2.0.

Good at

  • reasoning
  • tools usage
  • experimental local use

Limits

  • requires 9.9 GB memory at 8 k context
  • no vision support
  • no official Hikari endorsement

Best for

  • local inference on high‑VRAM GPUs
  • research prototyping
  • tool‑integrated LLM workflows

Who it’s for. Developers and researchers who need a large Qwen35 model for local experimentation. The model is not intended for production use and is a derivative of Hikari07jp, not an official release.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—7.2 GB 9.9 GB
Too big Get file →

The only available quant is the default GGUF file (7.21 GB). It needs roughly 9.9 GB of RAM at 8 k context. For 8 GB VRAM, the model cannot run. If a smaller quant existed, it would reduce memory, but none is provided.

What GPU do you need for Hikari Bonsai 27B Nimble LoRA?

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: the best quant that fits is the main file (7.2 GB file, about 9.9 GB in use).

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters26.9B
Layers64
Trained context262,144 tokens
KV cache (fp16)256 KB per token

Source

Quantized byzapabobouj
FamilyHikari
Licenseapache-2.0
Downloads (HF)278
Last updated2026-10-04

Frequently asked questions about Hikari Bonsai 27B Nimble LoRA

How much VRAM does Hikari Bonsai 27B Nimble LoRA need?

About 9.9 GB at its main file with an 8K context window. That is the 7.2 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 Hikari Bonsai 27B Nimble LoRA on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 9.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Hikari Bonsai 27B Nimble LoRA GGUF?

It depends on the quant: file is 7.2 GB.

What context length does Hikari Bonsai 27B Nimble LoRA support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.

Where can I download Hikari Bonsai 27B Nimble LoRA GGUF?

The files are on Hugging Face in the zapabobouj/Hikari-Bonsai-27B-Nimble-LoRA-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 Hikari Bonsai 27B Nimble LoRA 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 this model on 8 GB VRAM?

No. The model requires about 9.9 GB of memory at 8 k context, which exceeds 8 GB VRAM.

What license does the model use?

The model is released under apache‑2.0.

Is this an official Hikari release?

No. It is a derivative published by zapabobouj and not an official Hikari release.

Similar models

Browse all →

More from zapabobouj

All by zapabobouj →

More Hikari models

All Hikari →

Other medium models

All medium models →