Instructions to use Ttimms/MiniCPM5-2B-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Ttimms/MiniCPM5-2B-NVFP4 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Ttimms/MiniCPM5-2B-NVFP4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Ttimms/MiniCPM5-2B-NVFP4")
model = AutoModelForCausalLM.from_pretrained("Ttimms/MiniCPM5-2B-NVFP4", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Ttimms/MiniCPM5-2B-NVFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Ttimms/MiniCPM5-2B-NVFP4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ttimms/MiniCPM5-2B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Ttimms/MiniCPM5-2B-NVFP4

SGLang

How to use Ttimms/MiniCPM5-2B-NVFP4 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Ttimms/MiniCPM5-2B-NVFP4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ttimms/MiniCPM5-2B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Ttimms/MiniCPM5-2B-NVFP4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ttimms/MiniCPM5-2B-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Ttimms/MiniCPM5-2B-NVFP4 with Docker Model Runner:
```
docker model run hf.co/Ttimms/MiniCPM5-2B-NVFP4
```

MiniCPM5-2B — NVFP4 W4A16 (GPTQ)

NVFP4 weight-only (W4A16) quantization of openbmb/MiniCPM5-2B — 4-bit float weights in 16-element blocks with FP8 block scales, activations in bf16, lm_head + embeddings kept in bf16. Rounding is GPTQ (Hessian-aware), not RTN, which recovers most of the quality NVFP4 otherwise loses on a small model. Produced with llm-compressor 0.13.

2.03 GiB on disk (bf16 base is 4.68 GiB — 57 % smaller)
Serves on vLLM (compressed-tensors); on SM120 the weight-only path currently decodes via the Marlin kernel to a bf16 GEMM (no native FP4 compute), so the benefit here is footprint, not raw speed
Coding cost vs bf16: ~2.4 pp HumanEval / ~4.8 pp MBPP — see the eval

Pick this if you need to fit near 2 GB (e.g. an 8 GB card with a large KV cache). If you can spare ~0.8 GB, the FP8 build is near-lossless.

Evaluation

lm-evaluation-harness, vLLM 0.26 backend, greedy, median of 3 draws with the range (the harness is non-deterministic run-to-run even at greedy). All arms measured in one session against the released checkpoint. HumanEval-instruct pass@1/create_test (n = 164); MBPP base 3-shot (n = 500).

build	HumanEval-inst	MBPP (3-shot)	size	Δ HE / MBPP
bf16 base	86.59 % (85.98–86.59)	50.60 % (50.40–51.00)	4.68 GiB	—
FP8-dynamic	84.76 % (84.15–85.37)	48.80 % (48.80–49.00)	2.84 GiB	−1.8 / −1.8 pp
NVFP4-W4A16 GPTQ (this)	84.15 % (81.71–84.15)	45.80 % (45.60–46.40)	2.03 GiB	−2.4 / −4.8 pp
NVFP4-W4A16 RTN	79.88 % (77.44–79.88)	41.20 % (41.20–41.80)	2.03 GiB	−6.7 / −9.4 pp
mixed (MLP-NVFP4 + attn-FP8)	84.15 % (82.93–85.98)	46.60 % (46.00–46.80)	2.20 GiB	−2.4 / −4.0 pp

GPTQ rounding recovers ~4.3 pp (HumanEval) / ~4.6 pp (MBPP) over plain RTN. A 2.5 B dense model still can't fully absorb 4-bit weights — the residual gap, especially on MBPP, is the cost of the extra ~0.8 GB saved vs FP8. A mixed-precision build (4-bit MLP + 8-bit attention) was tried and gives no advantage — same HumanEval, 0.17 GiB larger — so it is not published.

Full comparison, serving notes, and a fine-tune experiment that regressed the base's coding (so these quants target the released checkpoint): https://github.com/t-timms/minicpm5-2b-quants

Usage

vllm serve Ttimms/MiniCPM5-2B-NVFP4 --max-model-len 32768 --kv-cache-dtype fp8

On WSL, set VLLM_USE_V2_MODEL_RUNNER=0 (the V2 runner needs UVA, which WSL disables).

Method & provenance

Quantizer: llm-compressor 0.13, GPTQModifier(scheme="NVFP4A16", dampening_frac=0.1), ignore lm_head + embed_tokens. Calibration: 512 samples of theblackcat102/evol-codealpaca-v1 (code, disjoint from the eval sets), 2048 tokens.
Base: openbmb/MiniCPM5-2B (LlamaForCausalLM, 2.5 B, Apache-2.0).
Built and evaluated on an RTX 5070 Ti (Blackwell, SM120), vLLM 0.26.

License

Apache-2.0, inherited from openbmb/MiniCPM5-2B.

Downloads last month: 1,571

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for Ttimms/MiniCPM5-2B-NVFP4

Base model

openbmb/MiniCPM5-2B

Quantized

(91)

this model