MI

ghostai pro

MiniCPMGGUF by immortaltatsuUpdated 2026-10-14
Download on Hugging Face
Parameters2.52B
Context128K
Layers42
Architecturellama
Smallest quant1.4 GB
Quick answer

ghostai pro needs about 2.4 GB of memory at Q4_K_M (a 1.6 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 1.4 GB (IQ4_XS) to 1.6 GB (Q4_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 1.4 GB.

GhostAI Pro – on‑device tool‑calling model based on MiniCPM5‑2B

GhostAI Pro is a 2.52B‑parameter model fine‑tuned from openbmb/MiniCPM5‑2B for the GhostWallet Solana app. It supports reasoning and tool use within a 131072‑token context. The model is released under Apache‑2.0 and is available in 4‑bit GGUF formats.

Good at

  • Tool‑calling with Hermes style output
  • Reasoning and math tasks
  • Long‑context understanding

Limits

  • No vision support
  • Requires 8‑k context for memory estimates
  • No mixture of experts

Best for

  • Solana app contract interactions
  • On‑device local deployment
  • Tool‑enabled LLM workflows

Who it’s for. Developers building Solana applications that need a lightweight, on‑device LLM for tool‑calling. The model is suitable for local deployment scenarios where privacy and low latency are priorities.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ4_XS1.4 GB 2.3 GB
Fits Get file →
Q4_K_M1.6 GB 2.4 GB
Fits Get file →

For 8‑GB VRAM, use IQ4_XS (1.42 GB, ~2.3 GB RAM at 8k context). For 16‑GB VRAM, Q4_K_M (1.56 GB, ~2.4 GB RAM at 8k context) offers slightly higher precision. Estimates are approximate.

What GPU do you need for ghostai pro?

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 Q4_K_M (1.6 GB file, about 2.4 GB in use).

Which ghostai pro quant should you choose?

Q4_K_M is the usual starting point for ghostai pro: it balances file size and quality well. Here it is a 1.6 GB file.

If memory is tight, the smallest file is IQ4_XS at 1.4 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters2.52B
Layers42
Trained context131,072 tokens
KV cache (fp16)42 KB per token

Source

Quantized byimmortaltatsu
FamilyMiniCPM
Licenseapache-2.0
Downloads (HF)990
Last updated2026-10-14

Frequently asked questions about ghostai pro

How much VRAM does ghostai pro need?

About 2.4 GB at Q4_K_M with an 8K context window. That is the 1.6 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 ghostai pro on an 8 GB GPU?

Yes, with Q4_K_M: the 1.6 GB file needs about 2.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of ghostai pro GGUF?

It depends on the quant: IQ4_XS is 1.4 GB, Q4_K_M is 1.6 GB.

What context length does ghostai pro support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.4 GB at 32K.

Where can I download ghostai pro GGUF?

The files are on Hugging Face in the immortaltatsu/ghostai-pro repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does ghostai pro 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 GhostAI Pro on 8 GB VRAM?

Yes, the IQ4_XS quant is 1.42 GB and requires about 2.3 GB RAM at 8k context, fitting within 8 GB VRAM.

Does GhostAI Pro support vision?

No, the model has no vision capabilities.

What license applies to GhostAI Pro?

The model is released under the Apache‑2.0 license.

Other GGUFs of MiniCPM5-2B

Similar models

Browse all →

More from immortaltatsu

All by immortaltatsu →

More MiniCPM models

All MiniCPM →

Other tiny models

All tiny models →