Model Architecture: Qwen2ForCausalLM
- Input: Text
- Output: Text
Model Optimizations:
- Weight quantization: FP8
- Activation quantization: FP8
Release Date: 2/5/2025
Version: 1.0
Model Developers: Neural Magic

Quantized version of DeepSeek-R1-Distill-Qwen-32B.

Model Optimizations

This model was obtained by quantizing the weights and activations of DeepSeek-R1-Distill-Qwen-32B to FP8 data type. This optimization reduces the number of bits per parameter from 16 to 8, reducing the disk size and GPU memory requirements by approximately 50%.

Only the weights and activations of the linear operators within transformers blocks are quantized. Weights are quantized using a symmetric per-channel scheme, whereas quantizations are quantized using a symmetric per-token scheme. LLM Compressor is used for quantization.

Use with vLLM

This model can be deployed efficiently using the vLLM backend, as shown in the example below.

python

from transformers import AutoTokenizer
from vllm import LLM, SamplingParams

number_gpus = 1
model_name = "neuralmagic/DeepSeek-R1-Distill-Qwen-32B-dynamic"

tokenizer = AutoTokenizer.from_pretrained(model_name)
sampling_params = SamplingParams(temperature=0.6, max_tokens=256, stop_token_ids=[tokenizer.eos_token_id])
llm = LLM(model=model_name, tensor_parallel_size=number_gpus, trust_remote_code=True)

messages_list = [
    [{"role": "user", "content": "Who are you? Please respond in pirate speak!"}],
]

prompt_token_ids = [tokenizer.apply_chat_template(messages, add_generation_prompt=True) for messages in messages_list]

outputs = llm.generate(prompt_token_ids=prompt_token_ids, sampling_params=sampling_params)

generated_text = [output.outputs[0].text for output in outputs]
print(generated_text)

vLLM also supports OpenAI-compatible serving. See the documentation for more details.

Creation

This model was created with llm-compressor by running the code snippet below.

python

from transformers import AutoModelForCausalLM, AutoTokenizer
from llmcompressor.modifiers.quantization import QuantizationModifier
from llmcompressor.transformers import oneshot
import os

# Load model
model_stub = "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B"
model_name = model_stub.split("/")[-1]

model = AutoModelForCausalLM.from_pretrained(
    model_stub,
    torch_dtype="auto",
)

tokenizer = AutoTokenizer.from_pretrained(model_stub)

# Configure the quantization algorithm and scheme
recipe = QuantizationModifier(
    targets="Linear",
    scheme="FP8_DYNAMIC",
    ignore=["lm_head"],
)

# Apply quantization
oneshot(
    model=model,
    recipe=recipe,
)

# Save to disk in compressed-tensors format
save_path = model_name + "-FP8-dynamic
model.save_pretrained(save_path)
tokenizer.save_pretrained(save_path)
print(f"Model and tokenizer saved to: {save_path}")

Evaluation

The model was evaluated on OpenLLM Leaderboard V1 and V2, using the following commands:

OpenLLM Leaderboard V1:

lm_eval \
  --model vllm \
  --model_args pretrained="neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-dynamic",dtype=auto,max_model_len=4096,tensor_parallel_size=1,enable_chunked_prefill=True \
  --tasks openllm \
  --write_out \
  --batch_size auto \
  --output_path output_dir \
  --show_config

OpenLLM Leaderboard V2:

lm_eval \
  --model vllm \
  --model_args pretrained="neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-dynamic",dtype=auto,max_model_len=4096,tensor_parallel_size=1,enable_chunked_prefill=True \
  --apply_chat_template \
  --fewshot_as_multiturn \
  --tasks leaderboard \
  --write_out \
  --batch_size auto \
  --output_path output_dir \
  --show_config

Accuracy

Category	Metric	deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-dynamic	Recovery
Reasoning	AIME 2024 (pass@1)	69.75	68.5	98.21%
MATH-500 (pass@1)	95.09	95.26	100.18%
GPQA Diamond (pass@1)	64.05	62.88	98.17%
Average Score	76.3	75.55	99.02%
OpenLLM V1	ARC-Challenge (Acc-Norm, 25-shot)	64.59	64.42	99.7%
GSM8K (Strict-Match, 5-shot)	82.71	82.64	99.9%
HellaSwag (Acc-Norm, 10-shot)	83.80	83.77	100.0%
MMLU (Acc, 5-shot)	81.12	80.98	99.8%
TruthfulQA (MC2, 0-shot)	58.41	58.30	99.8%
Winogrande (Acc, 5-shot)	76.40	76.09	99.6%
Average Score	74.51	74.36	99.8%
OpenLLM V2	IFEval (Inst Level Strict Acc, 0-shot)	42.87	49.43	99.2%
BBH (Acc-Norm, 3-shot)	57.96	58.38	100.7%
Math-Hard (Exact-Match, 4-shot)	0.00	0.00	---
GPQA (Acc-Norm, 0-shot)	26.95	26.86	99.7%
MUSR (Acc-Norm, 0-shot)	43.95	44.22	100.6%
MMLU-Pro (Acc, 5-shot)	49.82	49.43	99.2%
Average Score	36.92	36.86	99.8%
Coding	HumanEval (pass@1)	86.00	85.20	99.1%
HumanEval (pass@10)	92.50	92.20	99.7%
HumanEval+ (pass@10)	82.00	80.90	98.7%
HumanEval+ (pass@10)	88.70	88.70	100.0%

Inference Performance

This model achieves up to 1.5x speedup in single-stream deployment and up to 1.7x speedup in multi-stream asynchronous deployment, depending on hardware and use-case scenario. The following performance benchmarks were conducted with vLLM version 0.7.2, and GuideLLM.

Benchmarking Command

guidellm --model neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-dynamic --target "http://localhost:8000/v1" --data-type emulated --data "prompt_tokens=<prompt_tokens>,generated_tokens=<generated_tokens>" --max seconds 360 --backend aiohttp_server

Single-stream performance (measured with vLLM version 0.7.2)

				Instruction Following 256 / 128	Multi-turn Chat 512 / 256	Docstring Generation 768 / 128	RAG 1024 / 128	Code Completion 256 / 1024	Code Fixing 1024 / 1024	Large Summarization 4096 / 512	Large RAG 10240 / 1536
GPU class	Number of GPUs	Model	Average cost reduction	Latency (s)	QPD	Latency (s)	QPD	Latency (s)	QPD	Latency (s)	QPD	Latency (s)	QPD	Latency (s)	QPD	Latency (s)	QPD	Latency (s)	QPD
A6000	2	deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	---	6.3	359	12.8	176	6.5	347	6.6	342	49.9	45	50.8	44	26.6	85	83.4	27
1	neuralmagic/DeepSeek-R1-Distill-Qwen-32B-quantized.w8a8	1.81	6.9	648	13.8	325	7.2	629	7.2	622	54.8	82	55.6	81	30.0	150	94.8	47
1	neuralmagic/DeepSeek-R1-Distill-Qwen-32B-quantized.w4a16	3.07	3.9	1168	7.8	580	4.3	1041	4.6	975	29.7	151	30.9	146	19.3	233	61.4	73
A100	1	deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	---	5.6	361	11.1	180	5.7	350	5.8	347	44.0	46	44.7	45	23.6	85	73.7	27
1	neuralmagic/DeepSeek-R1-Distill-Qwen-32B-quantized.w8a8	1.50	3.7	547	7.3	275	3.8	536	3.8	528	29.0	69	29.5	68	15.7	128	53.1	38
1	neuralmagic/DeepSeek-R1-Distill-Qwen-32B-quantized.w4a16	2.30	2.2	894	4.5	449	2.4	831	2.5	798	17.4	116	18.0	112	10.5	191	49.5	41
H100	1	deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	---	3.3	327	6.7	163	3.4	320	3.4	317	26.6	41	26.9	41	14.3	77	47.8	23
1	neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-dynamic	1.52	2.2	503	4.3	252	2.2	490	2.3	485	17.3	63	17.5	63	9.5	116	33.4	33
1	neuralmagic/DeepSeek-R1-Distill-Qwen-32B-quantized.w4a16	1.61	2.1	532	4.1	268	2.1	516	2.1	513	16.1	68	16.5	66	9.1	120	31.9	34

**Use case profiles: prompt tokens / generation tokens

**QPD: Queries per dollar, based on on-demand cost at Lambda Labs (observed on 2/18/2025).

Multi-stream asynchronous performance (measured with vLLM version 0.7.2)

			Instruction Following 256 / 128	Multi-turn Chat 512 / 256	Docstring Generation 768 / 128	RAG 1024 / 128	Code Completion 256 / 1024	Code Fixing 1024 / 1024	Large Summarization 4096 / 512	Large RAG 10240 / 1536
Hardware	Model	Average cost reduction	Maximum throughput (QPS)	QPD	Maximum throughput (QPS)	QPD	Maximum throughput (QPS)	QPD	Maximum throughput (QPS)	QPD	Maximum throughput (QPS)	QPD	Maximum throughput (QPS)	QPD	Maximum throughput (QPS)	QPD	Maximum throughput (QPS)	QPD
A6000x2	deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	---	6.2	13940	1.9	4348	2.7	6153	2.1	4778	0.6	1382	0.4	930	0.3	685	0.1	124
neuralmagic/DeepSeek-R1-Distill-Qwen-32B-quantized.w8a8	1.80	8.7	19492	4.2	9474	4.1	9290	3.0	6802	1.2	2734	0.9	1962	0.5	1177	0.1	254
neuralmagic/DeepSeek-R1-Distill-Qwen-32B-quantized.w4a16	1.30	5.9	13366	2.5	5733	2.4	5409	1.6	3525	1.2	2757	0.7	1663	0.3	676	0.1	214
A100x2	deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	---	12.9	13016	5.8	5848	6.3	6348	5.1	5146	2.0	1988	1.5	1463	0.9	869	0.2	192
neuralmagic/DeepSeek-R1-Distill-Qwen-32B-quantized.w8a8	1.52	21.4	21479	8.9	8948	10.6	10611	8.2	8197	3.0	3018	2.0	2054	1.2	1241	0.3	264
neuralmagic/DeepSeek-R1-Distill-Qwen-32B-quantized.w4a16	1.09	13.5	13568	6.5	6509	6.0	6075	4.7	4754	2.8	2790	1.6	1651	0.9	862	0.2	225
H100x2	deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	---	25.5	14392	12.5	7035	14.0	7877	11.3	6364	3.6	2041	2.7	1549	1.9	1057	0.4	200
neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-dynamic	1.46	46.7	25538	20.3	11082	23.3	12728	18.4	10049	5.3	2881	3.7	2097	2.6	1445	0.5	256
neuralmagic/DeepSeek-R1-Distill-Qwen-32B-quantized.w4a16	1.23	36.9	20172	17.4	9500	18.0	9822	14.2	7755	5.3	2900	3.3	1867	2.3	1265	0.4	241

**Use case profiles: prompt tokens / generation tokens

**QPS: Queries per second.

**QPD: Queries per dollar, based on on-demand cost at Lambda Labs (observed on 2/18/2025).

RedHatAI/DeepSeek-R1-Distill-Qwen-32B-FP8-dynamic

DeepSeek-R1-Distill-Qwen-32B-FP8-dynamic

Model Overview