File size: 4,814 Bytes
7f8a0b1 f5e5c49 7f8a0b1 9dabd59 7f8a0b1 9dabd59 7f8a0b1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 |
---
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-72B/blob/main/LICENSE
language:
- en
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-72B
tags:
- chat
- neuralmagic
- llmcompressor
---
# Qwen2.5-72B-quantized.w8a8
## Model Overview
- **Model Architecture:** Qwen2
- **Input:** Text
- **Output:** Text
- **Model Optimizations:**
- **Activation quantization:** INT8
- **Weight quantization:** INT8
- **Intended Use Cases:** Intended for commercial and research use multiple languages. Similarly to [Qwen2.5-72B](https://huggingface.co/Qwen/Qwen2.5-72B), this models is intended for assistant-like chat.
- **Out-of-scope:** Use in any manner that violates applicable laws or regulations (including trade compliance laws).
- **Release Date:** 10/09/2024
- **Version:** 1.0
- **Model Developers:** Neural Magic
Quantized version of [Qwen2.5-72B](https://huggingface.co/Qwen/Qwen2.5-72B).
It achieves an average score of 78.71 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 78.91.
### Model Optimizations
This model was obtained by quantizing the weights and activations of [Qwen2.5-72B](https://huggingface.co/Qwen/Qwen2.5-72B) to INT8 data type.
This optimization reduces the number of bits used to represent weights and activations from 16 to 8, reducing GPU memory requirements (by approximately 50%) and increasing matrix-multiply compute throughput (by approximately 2x).
Weight quantization also reduces disk size requirements by approximately 50%.
Only weights and activations of the linear operators within transformers blocks are quantized.
Weights are quantized with a symmetric static per-channel scheme, where a fixed linear scaling factor is applied between INT8 and floating point representations for each output channel dimension.
Activations are quantized with a symmetric dynamic per-token scheme, computing a linear scaling factor at runtime for each token between INT8 and floating point representations.
## Deployment
This model can be deployed efficiently using the [vLLM](https://docs.vllm.ai/en/latest/) backend, as shown in the example below.
```python
from vllm import LLM, SamplingParams
from transformers import AutoTokenizer
model_id = "neuralmagic-ent/Qwen2.5-72B-quantized.w8a8"
number_gpus = 2
max_model_len = 8192
sampling_params = SamplingParams(temperature=0.7, top_p=0.8, max_tokens=256)
tokenizer = AutoTokenizer.from_pretrained(model_id)
prompt = "Give me a short introduction to large language model."
llm = LLM(model=model_id, tensor_parallel_size=number_gpus, max_model_len=max_model_len)
outputs = llm.generate(prompt, sampling_params)
generated_text = outputs[0].outputs[0].text
print(generated_text)
```
vLLM aslo supports OpenAI-compatible serving. See the [documentation](https://docs.vllm.ai/en/latest/) for more details.
## Evaluation
The model was evaluated on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) leaderboard tasks (version 1) with the [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness/tree/3872Bbd54bc621086e05aa1b030d8d4d5635b25e6) (commit 3872Bbd54bc621086e05aa1b030d8d4d5635b25e6) and the [vLLM](https://docs.vllm.ai/en/stable/) engine, using the following command:
```
lm_eval \
--model vllm \
--model_args pretrained="neuralmagic-ent/Qwen2.5-72B-quantized.w8a8",dtype=auto,gpu_memory_utilization=0.9,add_bos_token=True,max_model_len=4096,enable_chunk_prefill=True,tensor_parallel_size=2 \
--tasks openllm \
--batch_size auto
```
### Accuracy
#### Open LLM Leaderboard evaluation scores
<table>
<tr>
<td><strong>Benchmark</strong>
</td>
<td><strong>Qwen2.5-72B</strong>
</td>
<td><strong>Qwen2.5-72B-quantized.w8a8 (this model)</strong>
</td>
<td><strong>Recovery</strong>
</td>
</tr>
<tr>
<td>MMLU (5-shot)
</td>
<td>86.05
</td>
<td>85.68
</td>
<td>99.6%
</td>
</tr>
<tr>
<td>ARC Challenge (25-shot)
</td>
<td>68.26
</td>
<td>68.17
</td>
<td>99.9%
</td>
</tr>
<tr>
<td>GSM-8K (5-shot, strict-match)
</td>
<td>88.63
</td>
<td>88.63
</td>
<td>100.0%
</td>
</tr>
<tr>
<td>Hellaswag (10-shot)
</td>
<td>87.53
</td>
<td>87.53
</td>
<td>100.0%
</td>
</tr>
<tr>
<td>Winogrande (5-shot)
</td>
<td>82.56
</td>
<td>81.77
</td>
<td>99.0%
</td>
</tr>
<tr>
<td>TruthfulQA (0-shot, mc2)
</td>
<td>60.44
</td>
<td>60.49
</td>
<td>100.1%
</td>
</tr>
<tr>
<td><strong>Average</strong>
</td>
<td><strong>78.91</strong>
</td>
<td><strong>78.71</strong>
</td>
<td><strong>99.8%</strong>
</td>
</tr>
</table>
|