VI

RefinedToolCallV5 3b

VibethinkerGGUF by RefinedNeuroUpdated 2026-06-27
Download on Hugging Face
Parameters3.09B
Context128K
Layers36
Architectureqwen2
Smallest quant2.5 GB
Quick answer

RefinedToolCallV5 3b needs about 4.1 GB of memory at Q8_0 (a 3.3 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 3 quants, from 2.5 GB (Q6_K) to 6.2 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.2 GB.

3B parameter model that reasons and calls tools in multi‑turn conversations

RefinedToolCallV5 3b is a 3.09B parameter model built on the Qwen2 architecture. It supports reasoning and tool‑calling with a 131072 token context window. The model is released under Apache‑2.0 and is available in several GGUF quantizations.

Good at

  • Multi‑turn tool‑calling
  • Reasoning with verification
  • Low‑memory local deployment

Limits

  • No vision support
  • Requires 8k+ context for full performance
  • No GPU required but benefits from it

Best for

  • Local agentic applications
  • Math‑grade reasoning tasks
  • Function‑calling workflows

Who it’s for. Developers needing a small, local model that can handle multi‑turn tool interactions and reasoning. The model is suitable for edge deployments and fine‑tuning on specialized data sets.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q6_K2.5 GB 3.3 GB
Fits Get file →
Q8_03.3 GB 4.1 GB
Fits Get file →
F166.2 GB 7.0 GB
Fits Get file →

For 8 GB VRAM, use the Q6_K quantization (2.54 GB file, ~3.3 GB memory at 8k context). If more VRAM is available, Q8_0 (3.29 GB file, ~4.1 GB memory) offers higher precision. The F16 quant is only suitable for systems with ≥7 GB memory at 8k context.

What GPU do you need for RefinedToolCallV5 3b?

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 Q6_K (2.5 GB file, about 3.3 GB in use).
  • 6 GB VRAM: the best quant that fits is Q8_0 (3.3 GB file, about 4.1 GB in use).
  • 8 GB VRAM: the best quant that fits is F16 (6.2 GB file, about 7.0 GB in use).

Which RefinedToolCallV5 3b quant should you choose?

Q8_0 (3.3 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.

If memory is tight, the smallest file is Q6_K at 2.5 GB, with the quality trade-offs that come with fewer bits. Very close to the original quality for most uses, at a clearly larger size than 4-bit.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen2
Parameters3.09B
Layers36
Trained context131,072 tokens
KV cache (fp16)36 KB per token

Source

Quantized byRefinedNeuro
Licenseapache-2.0
Downloads (HF)885
Last updated2026-06-27

Frequently asked questions about RefinedToolCallV5 3b

How much VRAM does RefinedToolCallV5 3b need?

About 4.1 GB at Q8_0 with an 8K context window. That is the 3.3 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 RefinedToolCallV5 3b on an 8 GB GPU?

Yes, with F16: the 6.2 GB file needs about 7.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of RefinedToolCallV5 3b GGUF?

It depends on the quant: Q6_K is 2.5 GB, Q8_0 is 3.3 GB, F16 is 6.2 GB.

What context length does RefinedToolCallV5 3b support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.3 GB at 8K and 1.2 GB at 32K.

Where can I download RefinedToolCallV5 3b GGUF?

The files are on Hugging Face in the RefinedNeuro/RefinedToolCallV5-3b repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does RefinedToolCallV5 3b 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?

Yes, the Q6_K quantization requires about 3.3 GB of VRAM at 8k context, which fits within an 8 GB budget.

Does this model support vision?

No, the model does not have vision capabilities.

Is it capable of multi‑turn conversations?

Yes, it is designed for multi‑turn tool‑calling and agentic interactions.

Other GGUFs of VibeThinker-3B

Similar models

Browse all →

More from RefinedNeuro

All by RefinedNeuro →

More Vibethinker models

All Vibethinker →

Other tiny models

All tiny models →