MiniCPM5-2B NVFP4 GGUF
GGUF quantized version of openbmb/MiniCPM5-2B, the second model in the MiniCPM5 series. It is a dense 2B Transformer (LlamaForCausalLM) built for on-device, local deployment, and resource-constrained scenarios, reaching 2B-class open-source SOTA.
Model Overview
MiniCPM5-2B is designed for local assistants, coding agents, tool-use workflows, and reasoning scenarios where a compact model is preferred. It keeps a small deployment footprint while providing native long-context support (131,072 tokens), and was post-trained with RL + OPD (On-Policy Distillation). It supports a chat template with thinking/reasoning mode (controlled via enable_thinking) and XML-style tool calling.
Model Architecture
| Property | Value |
|---|---|
| Architecture | LlamaForCausalLM (dense) |
| Total Parameters | 2.52B |
| Non-Embedding Parameters | 1.98B |
| Layers | 42 |
| Attention Heads (GQA) | 16 Q / 2 KV |
| Context Length | 131,072 (128K) |
| Original Precision | bfloat16 |
| Supported Languages | en, zh |
Available GGUF Files
| File | Quantization | Size | Use Case |
|---|---|---|---|
minicpm5-2b-nvfp4.gguf | NVFP4 | 1.4 GB | 4-bit NVFP4, compact, quality/size balance |
minicpm5-2b-dspark-nvfp4.gguf | NVFP4 | 1.4 GB | Draft model |
Usage
llama.cpp CLI
./llama-cli \
-m minicpm5-2b-nvfp4.gguf \
-p "Explain quantum computing in simple terms." \
--temp 1.0 --top-p 0.95
llama-server (OpenAI-compatible API)
./llama-server \
-m minicpm5-2b-nvfp4.gguf \
--host 0.0.0.0 --port 8080
The GGUF also works with Ollama and LM Studio.
Thinking Mode
The model supports deep-thinking output. You can control it per request via the chat template, e.g. with an OpenAI-compatible API:
"chat_template_kwargs": {"enable_thinking": false}
Tool Calling
MiniCPM5-2B emits XML-style tool calls. Tool definitions are injected into the prompt, and tool results are returned in the observation/tool role. SGLang's built-in minicpm5 parser converts these to OpenAI-compatible tool_calls natively (see upstream model card).
Generation Parameters
Recommended parameters from the original model:
| Parameter | Value |
|---|---|
| Temperature | 1.0 |
| Top-P | 0.95 |
Quantization
This GGUF was quantized to NVFP4 (a 4-bit floating-point format) from the BF16 source model, verified with llama.cpp's GGUF reader. All weight tensors use NVFP4; the embedding/output and norm tensors are kept in higher precision (Q6_K / F32) to preserve quality. Recommended sampling settings are stored in the file's metadata: temperature 1.0, top-p 0.95.
Acknowledgements
- Original model: openbmb/MiniCPM5-2B
- Quantization tool: llama.cpp