Modelpublic

PollardWeights/MiniCPM5-2B-Pollard

sourceHugging Faceapache-2.0updated 16d agoView on Hugging Face

1likes746downloads

Model Card

MiniCPM5-2B — Pollard

### Pollard shrank this model: 5.04 GB (f16) → 1.41 GB — 72% smaller, 3.6× down. The smallest rung here; larger, higher-fidelity rungs are listed below. | format | this model's size | |---|---:| | f16 | 5.04 GB | | Q80 | ~2.67 GB | | Q6K | ~2.07 GB | | Q4KM | ~1.46 GB | | PollardMix (this repo's IQ4_XS) | 1.41 GB |

Pollard builds of openbmb/MiniCPM5-2B made with Pollard Weights — a ladder of measured-allocation quants (bits placed by per-layer sensitivity, not a uniform crush).

Standard GGUF — runs in stock llama.cpp / ik_llama.cpp, Ollama, LM Studio. Trellis (IQ*_KT) files need ik_llama.cpp; the K-quants run anywhere.

Model details


Parameter count	~2.5B
Architecture	`llama`
Input support	text
imatrix	no
Perplexity measured	yes — table below

Which file should I choose?

Every rung is the same weights, sized to a different RAM budget by the measured allocation. Pick the largest one that fits your machine with room for context:

—~4 GB RAM / VRAM → `Q6_K` (2.07 GB). Q6_K
—~4 GB RAM / VRAM → `Q5_K_M` (1.74 GB). Q5KM
—~3 GB RAM / VRAM → `IQ4_XS` (1.41 GB). IQ4_XS

Available files

file	PPL	size	Mean KLD	notes
`MiniCPM5-2B-Pollard-IQ4_XS.gguf`	—	1.41 GB	—	IQ4_XS
`MiniCPM5-2B-Pollard-Q5_K_M.gguf`	—	1.74 GB	—	Q5KM
`MiniCPM5-2B-Pollard-Q6_K.gguf`	—	2.07 GB	—	Q6_K

Download a specific file

bash

pip install -U "huggingface_hub[cli]"
hf download PollardWeights/MiniCPM5-2B-Pollard \
  --include "MiniCPM5-2B-Pollard-IQ4_XS.gguf" --local-dir ./

How to run

These are standard GGUF and run with llama.cpp:

bash

llama-server -hf PollardWeights/MiniCPM5-2B-Pollard:IQ4_XS

or from a local file:

bash

llama-cli    -m MiniCPM5-2B-Pollard-IQ4_XS.gguf -ngl 99 -p "Explain why the sky is blue."
llama-server -m MiniCPM5-2B-Pollard-IQ4_XS.gguf -ngl 99      # OpenAI-compatible API + web UI at :8080

They also work in anything built on llama.cpp — LM Studio, koboldcpp, Jan, ramalama, Ollama (ollama run hf.co/PollardWeights/MiniCPM5-2B-Pollard).

ARM / AVX

llama.cpp repacks weights into an interleaved layout at load time for faster inference on ARM and AVX machines — no special file needed, online repacking covers these quants. The old Q4_0_4_4/4_8/8_8 variants are not required.

Errata

—Trellis (IQ*_KT) quants need ik_llama.cpp to build/run; K-quants run in any recent llama.cpp.
—Measured allocation places bits by per-layer sensitivity under a size budget.
—Single machine; replication invited.

Credits & license

—Base model: `openbmb/MiniCPM5-2B`
—Quantization tooling: llama.cpp (ggml-org)
—Method + tooling: Pollard Weights — measure first, no claim before a number.
—License: apache-2.0, inherited from the base model.

Built with [Pollard Weights](https://github.com/WestWaters/pollard-weights) — frontier models, small hardware, no compromise.