Create README.md

README.md

@@ -0,0 +1,174 @@
+---
+license: mit
+tags:
+- deepseek
+- fp8
+- vllm
+base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
+library_name: transformers
+---
+
+# DeepSeek-R1-Distill-Qwen-32B-FP8-Dynamic
+
+## Model Overview
+- **Model Architecture:** DeepSeek-R1-Distill-Qwen-32B
+  - **Input:** Text
+  - **Output:** Text
+- **Model Optimizations:**
+  - **Weight quantization:** FP8
+  - **Activation quantization:** FP8
+- **Release Date:** 2/6/2025
+- **Version:** 1.0
+- **Model Developers:** Neural Magic
+
+Quantized version of [DeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B).
+
+### Model Optimizations
+
+This model was obtained by quantizing the weights and activations to FP8 data type, ready for inference with vLLM.
+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. 
+
+## Deployment
+
+### Use with vLLM
+
+This model can be deployed efficiently using the [vLLM](https://docs.vllm.ai/en/latest/) backend, as shown in the example below.
+
+```python
+from transformers import AutoTokenizer
+from vllm import LLM, SamplingParams
+
+max_model_len, tp_size = 4096, 1
+model_name = "neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-Dynamic"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+llm = LLM(model=model_name, tensor_parallel_size=tp_size, max_model_len=max_model_len, trust_remote_code=True)
+sampling_params = SamplingParams(temperature=0.3, max_tokens=256, stop_token_ids=[tokenizer.eos_token_id])
+
+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](https://docs.vllm.ai/en/latest/) for more details.
+
+## Creation
+
+This model was created with [llm-compressor](https://github.com/vllm-project/llm-compressor) by running the code snippet below. 
+
+
+```python
+import argparse
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from llmcompressor.modifiers.quantization import QuantizationModifier
+from llmcompressor.transformers import oneshot
+import os
+
+def main():
+    parser = argparse.ArgumentParser(description='Quantize a transformer model to FP8')
+    parser.add_argument('--model_id', type=str, required=True,
+                        help='The model ID from HuggingFace (e.g., "meta-llama/Meta-Llama-3-8B-Instruct")')
+    parser.add_argument('--save_path', type=str, default='.',
+                        help='Custom path to save the quantized model. If not provided, will use model_name-FP8-dynamic')
+    args = parser.parse_args()
+
+    # Load model
+    model = AutoModelForCausalLM.from_pretrained(
+        args.model_id, device_map="auto", torch_dtype="auto", trust_remote_code=True,
+    )
+    tokenizer = AutoTokenizer.from_pretrained(args.model_id)
+
+    # Configure the quantization algorithm and scheme
+    recipe = QuantizationModifier(
+        targets="Linear", scheme="FP8_DYNAMIC", ignore=["lm_head"]
+    )
+
+    # Apply quantization
+    oneshot(model=model, recipe=recipe)
+
+    save_path = os.path.join(args.save_path, args.model_id.split("/")[1] + "-FP8-dynamic")
+    os.makedirs(save_path, exist_ok=True)
+
+    # Save to disk in compressed-tensors format
+    model.save_pretrained(save_path)
+    tokenizer.save_pretrained(save_path)
+    print(f"Model and tokenizer saved to: {save_path}")
+
+if __name__ == "__main__":
+    main()
+```
+
+## Evaluation
+
+The model was evaluated on OpenLLM Leaderboard [V1](https://huggingface.co/spaces/open-llm-leaderboard-old/open_llm_leaderboard) and [V2](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/), using the following commands:
+
+OpenLLM Leaderboard V1:
+```
+lm_eval \
+  --model vllm \
+  --model_args pretrained="neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-Dynamic",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=1,gpu_memory_utilization=0.8,enable_chunked_prefill=True,trust_remote_code=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,add_bos_token=False,max_model_len=4096,tensor_parallel_size=1,gpu_memory_utilization=0.8,enable_chunked_prefill=True,trust_remote_code=True \
+  --apply_chat_template \
+  --fewshot_as_multiturn \
+  --tasks leaderboard \
+  --write_out \
+  --batch_size auto \
+  --output_path output_dir \
+  --show_config
+
+```
+
+### Accuracy
+
+#### OpenLLM Leaderboard V1 evaluation scores
+
+| Metric                                   | deepseek-ai/DeepSeek-R1-Distill-Qwen-32B             | neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-Dynamic |
+|-----------------------------------------|:---------------------------------:|:-------------------------------------------:|
+| ARC-Challenge (Acc-Norm, 25-shot)       |        64.59                     |                64.42                        |
+| GSM8K (Strict-Match, 5-shot)            |        82.71                     |                 82.64                        |
+| HellaSwag (Acc-Norm, 10-shot)           |        83.80                     |                     83.77                   |
+| MMLU (Acc, 5-shot)                      |        81.12                    |                       80.98              |
+| TruthfulQA (MC2, 0-shot)                |        58.41                     |                         58.30             |
+| Winogrande (Acc, 5-shot)                |       76.40                      |                 76.09                       |
+| **Average Score**                       | **74.51**                        | **74.36**                                   |
+| **Recovery (%)**                            | **100.00**                       | **99.79**                                   |
+
+#### OpenLLM Leaderboard V2 evaluation scores
+
+| Metric                                                   | deepseek-ai/DeepSeek-R1-Distill-Qwen-32B             | neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-Dynamic |
+|---------------------------------------------------------|:---------------------------------:|:-------------------------------------------:|
+| IFEval (Inst-and-Prompt Level Strict Acc, 0-shot)       |         42.87                    |               42.26                         |
+| BBH (Acc-Norm, 3-shot)                                  |         57.96                    |               58.38                         |
+| GPQA (Acc-Norm, 0-shot)                                 |         26.95                     |              26.86                           |
+| MUSR (Acc-Norm, 0-shot)                                 |         43.95                     |               44.22                         |
+| MMLU-Pro (Acc, 5-shot)                                  |     49.82                        |                49.43                        |
+| **Average Score**                                       | **44.31**                        | **44.23**                                   |
+| **Recovery (%)**                                            | **100.00**                       | **99.82**                                   |
+
+#### Coding evaluation scores
+
+| Metric                                                   | deepseek-ai/DeepSeek-R1-Distill-Qwen-32B             | neuralmagic/DeepSeek-R1-Distill-Qwen-32B-FP8-Dynamic |
+|---------------------------------------------------------|:---------------------------------:|:-------------------------------------------:|
+| HumanEval pass@1                                         |  86.00                           |      85.20                                  |
+| HumanEval pass@10                                        |  92.50                           |      92.20                                  |
+| HumanEval+ pass@1                                        |  82.00                           |      80.90                                  |
+| HumanEval+ pass@10                                       |  88.70                           |      88.70                                  |
+| **Average Score**                                       | **87.30**                        | **86.75**                                   |
+| **Recovery (%)**                                            | **100.00**                       | **99.37**                                   |