Instructions to use sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4 with Transformers:

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

pipe = pipeline("text-generation", model="sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4")
model = AutoModelForCausalLM.from_pretrained("sakamakismile/llm-jp-4-32b-a3b-thinking-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 sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sakamakismile/llm-jp-4-32b-a3b-thinking-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": "sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4

SGLang

How to use sakamakismile/llm-jp-4-32b-a3b-thinking-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 "sakamakismile/llm-jp-4-32b-a3b-thinking-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": "sakamakismile/llm-jp-4-32b-a3b-thinking-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 "sakamakismile/llm-jp-4-32b-a3b-thinking-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": "sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4 with Docker Model Runner:
```
docker model run hf.co/sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4
```

llm-jp-4-32b-a3b-thinking-NVFP4

NVFP4 (W4A4) quantization of llm-jp/llm-jp-4-32b-a3b-thinking — a Japanese-native 32B-A3B Mixture-of-Experts thinking (reasoning) model. Quantized with llm-compressor 0.12.0 to the compressed-tensors nvfp4-pack-quantized format, ready to serve on NVIDIA Blackwell (SM120) GPUs via vLLM with the fast FLASHINFER_CUTLASS MoE kernel + CUDA graphs.

~20.7 GB (vs ~64 GB bf16). Experts NVFP4 4-bit; lm_head kept in bf16; router/gates excluded.
Reasoning survives quantization and Japanese stays fluent; in internal testing it also writes look-ahead-safe trading/backtest code (uses .shift(1) for next-bar execution).

⚠️ Important: padded MoE intermediate (960 → 1024)

The base model's moe_intermediate_size is 960, which is not accepted by vLLM's fast NVFP4 MoE kernels (FLASHINFER_CUTLASS/MARLIN require 128-aligned, and 960/TP=480 is rejected).

This checkpoint therefore has each expert's intermediate dimension zero-padded from 960 to 1024 before quantization, and config.json reports moe_intermediate_size: 1024. The padding is mathematically lossless (the 64 extra units have zero gate/up output rows and zero down-projection columns, contributing exactly 0), but the tensor shapes differ from the base model — keep this in mind if you diff against llm-jp/llm-jp-4-32b-a3b-thinking.

Serving (vLLM ≥ 0.22, Blackwell)

vllm serve sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4 \
  --trust-remote-code \
  --tensor-parallel-size 2 \           # or 4
  --max-model-len 16384 \
  --kv-cache-dtype fp8 \
  --gpu-memory-utilization 0.90
# multi-GPU without NVLink/P2P: add  --disable-custom-all-reduce  and env NCCL_P2P_DISABLE=1

compressed-tensors quantization is auto-detected; FLASHINFER_CUTLASS NVFP4 MoE backend is selected automatically on Blackwell.
Reasoning format: the model uses the harmony channel format (analysis → final). vLLM 0.22's openai_gptoss reasoning parser raises NotImplementedError in non-streaming mode, so run without --reasoning-parser — the harmony text appears in content and the user-facing answer follows the final channel marker. (Use streaming if you want the parser to split reasoning_content.)
A custom tokenizer (llmjp4_tokenizer.py / llmjp4_harmony.py) ships with the model and loads via --trust-remote-code.

Measured throughput

On 2× / 4× NVIDIA RTX PRO 2000 Blackwell (16 GB, 288 GB/s, no NVLink), kv fp8, CUDA graphs:

	single-stream	aggregate
TP=2	~119 tok/s	~295 tok/s (4-way)
TP=4	~151–185 tok/s	~465 (4-way) / ~750 (8-way) tok/s

Quantization recipe

QuantizationModifier(targets="Linear", scheme="NVFP4", ignore=["lm_head", "re:.*mlp.gate$", "re:.*mlp.shared_expert_gate$"]), calibrated on neuralmagic/calibration (LLM split). MoE experts are linearized for per-expert calibration (calibrate_all_experts).

License & attribution

Apache-2.0, inherited from the base model llm-jp/llm-jp-4-32b-a3b-thinking by LLM-jp. Please cite and follow the base model's terms. This is a community quantization; no affiliation with LLM-jp.

Downloads last month: 89

Safetensors

Model size

20B params

Tensor type

F32

BF16

F8_E4M3

Model tree for sakamakismile/llm-jp-4-32b-a3b-thinking-NVFP4

Base model

llm-jp/llm-jp-4-32b-a3b-thinking

Quantized

(8)

this model