GE

SemanticRepair 270M

GemmaGGUF by Gramscii-ITUpdated 2026-10-05
Download on Hugging Face
Parameters268M
Context32K
Layers18
Architecturegemma3
Smallest quant0.3 GB
Quick answer

SemanticRepair 270M needs about 0.9 GB of memory at Q8_0 (a 0.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 0.6 GB.

SemanticRepair 270M – a routing‑only Gemma model

SemanticRepair 270M is a 270M‑parameter Gemma model that classifies request lines for an embedding router. It does not generate answers and relies on the host application for capability admission and tool execution.

Good at

  • Request‑line classification
  • Routing‑fidelity validation
  • Non‑answer generation

Limits

  • No vision support
  • No answer generation
  • Limited to routing tasks

Best for

  • Embedding router integration
  • Tool‑execution gating
  • Request‑presence classification

Who it’s for. Developers building routing layers for LLMs who need a lightweight model that only classifies requests. The model is suitable for systems that separate routing from generation.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_00.3 GB 0.9 GB
Fits Get file →

The Q8_0 quant is 0.29 GB and requires about 0.9 GB of VRAM at 8 k context. For 8 GB VRAM, Q8_0 fits comfortably. No other quants are available, so use Q8_0 for all memory budgets.

What GPU do you need for SemanticRepair 270M?

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: the best quant that fits is Q8_0 (0.3 GB file, about 0.9 GB in use).

Which SemanticRepair 270M quant should you choose?

Q8_0 (0.3 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturegemma3
Parameters268M
Layers18
Trained context32,768 tokens
KV cache (fp16)18 KB per token

Source

Quantized byGramscii-IT
FamilyGemma
Licensegemma
Downloads (HF)429
Last updated2026-10-05

Frequently asked questions about SemanticRepair 270M

How much VRAM does SemanticRepair 270M need?

About 0.9 GB at Q8_0 with an 8K context window. That is the 0.3 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 SemanticRepair 270M on an 8 GB GPU?

Yes, with Q8_0: the 0.3 GB file needs about 0.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of SemanticRepair 270M GGUF?

It depends on the quant: Q8_0 is 0.3 GB.

What context length does SemanticRepair 270M support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.6 GB at 32K.

Where can I download SemanticRepair 270M GGUF?

The files are on Hugging Face in the Gramscii-IT/SemanticRepair-270M repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does SemanticRepair 270M use?

Hugging Face lists the license as gemma. 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 SemanticRepair 270M on 8 GB VRAM?

Yes, the Q8_0 quant uses about 0.9 GB of VRAM at 8 k context, leaving room for other processes.

Does this model generate answers?

No, it only classifies request lines and does not produce answer text.

What context length does it support?

It was trained on 32,768 tokens and can handle that context length in inference.

Other GGUFs of gemma-3-270m

Similar models

Browse all →

More from Gramscii-IT

All by Gramscii-IT →

More Gemma models

All Gemma →

Other tiny models

All tiny models →