--- title: SmolLM-1.7B-Instruct-quantized.w8a16 canonical_url: "https://www.modelscope.cn/models/neuralmagic/SmolLM-1.7B-Instruct-quantized.w8a16" md_url: "https://www.modelscope.cn/models/neuralmagic/SmolLM-1.7B-Instruct-quantized.w8a16.md" repository: neuralmagic/SmolLM-1.7B-Instruct-quantized.w8a16 last_updated: 2026-08-24 license: apache-2.0 pipeline_tag: text-generation tasks: - text-generation model_type: - llama architectures: - LlamaForCausalLM base_model: - HuggingFaceTB/SmolLM-1.7B-Instruct base_model_relation: quantized parameters: 1.8B tensor_type: - I32 - I64 - BF16 library_name: - transformer - safetensors - pytorch frameworks: - pytorch language: - en downloads: 1990 stars: 0 tags: - int8 - vllm --- # SmolLM-1.7B-Instruct-quantized.w8a16 > SmolLM-1.7B-Instruct-quantized.w8a16 - neuralmagic 在 ModelScope 开源的模型。SmolLM-1.7B-Instruct-quantized.w8a16 neuralmagic/SmolLM-1.7B-Instruct-quantized.w8a16 是 ModelScope 魔搭社区上的 1.8B 参数text-generation模型，采用 apache-2.0 许可，基于 HuggingFaceTB/SmolLM-1.7B-Instruct 构建。 - **Repository**: neuralmagic/SmolLM-1.7B-Instruct-quantized.w8a16 - **License**: apache-2.0 - **Tasks**: text-generation - **Parameters**: 1.8B - **Base model**: HuggingFaceTB/SmolLM-1.7B-Instruct - **Tags**: int8, vllm - **Downloads**: 1990 - **Stars**: 0 - **Last updated**: 2026-08-24 Source: https://www.modelscope.cn/models/neuralmagic/SmolLM-1.7B-Instruct-quantized.w8a16 --- # SmolLM-1.7B-Instruct-quantized.w8a16 ## Model Overview - **Model Architecture:** Llama - **Input:** Text - **Output:** Text - **Model Optimizations:** - **Weight quantization:** INT8 - **Intended Use Cases:** Intended for commercial and research use in English. Similarly to [SmolLM-1.7B-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM-1.7B-Instruct), 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). Use in languages other than English. - **Release Date:** 8/16/2024 - **Version:** 1.0 - **License(s):** [Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0) - **Model Developers:** Neural Magic Quantized version of [SmolLM-1.7B-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM-1.7B-Instruct). It achieves an average score of 41.83 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 41.76. ### Model Optimizations This model was obtained by quantizing the weights of [SmolLM-1.7B-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM-1.7B-Instruct) to INT8 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 of the linear operators within transformers blocks are quantized. Symmetric per-channel quantization is applied, in which a linear scaling per output dimension maps the INT8 and floating point representations of the quantized weights. The [GPTQ](https://arxiv.org/abs/2210.17323) algorithm is applied for quantization, as implemented in the [llm-compressor](https://github.com/vllm-project/llm-compressor) library. GPTQ used a 1% damping factor and 1,024 sequences of 2,048 random tokens. ## 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 vllm import LLM, SamplingParams from transformers import AutoTokenizer model_id = "neuralmagic/SmolLM-1.7B-Instruct-quantized.w8a16" sampling_params = SamplingParams(temperature=0.6, top_p=0.92, max_tokens=100) tokenizer = AutoTokenizer.from_pretrained(model_id) messages = [ {"role": "user", "content": "List the steps to bake a chocolate cake from scratch."}, ] prompts = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) llm = LLM(model=model_id) outputs = llm.generate(prompts, sampling_params) generated_text = outputs[0].outputs[0].text 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 by using the [llm-compressor](https://github.com/vllm-project/llm-compressor) library as presented in the code snipet below. ```python from transformers import AutoTokenizer from datasets import Dataset from llmcompressor.transformers import SparseAutoModelForCausalLM, oneshot from llmcompressor.modifiers.quantization import GPTQModifier import random model_id = "HuggingFaceTB/SmolLM-1.7B-Instruct" num_samples = 1024 max_seq_len = 2048 tokenizer = AutoTokenizer.from_pretrained(model_id) max_token_id = len(tokenizer.get_vocab()) - 1 input_ids = [[random.randint(0, max_token_id) for _ in range(max_seq_len)] for _ in range(num_samples)] attention_mask = num_samples * [max_seq_len * [1]] ds = Dataset.from_dict({"input_ids": input_ids, "attention_mask": attention_mask}) recipe = GPTQModifier( targets="Linear", scheme="W8A16", ignore=["lm_head"], dampening_frac=0.01, ) model = SparseAutoModelForCausalLM.from_pretrained( model_id, device_map="auto", ) oneshot( model=model, dataset=ds, recipe=recipe, max_seq_length=max_seq_len, num_calibration_samples=num_samples, ) model.save_pretrained("SmolLM-1.7B-Instruct-quantized.w8a16") ``` ## 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/383bbd54bc621086e05aa1b030d8d4d5635b25e6) (commit 383bbd54bc621086e05aa1b030d8d4d5635b25e6) and the [vLLM](https://docs.vllm.ai/en/stable/) engine, using the following command: ``` lm_eval \ --model vllm \ --model_args pretrained="neuralmagic/SmolLM-1.7B-Instruct-quantized.w8a16",dtype=auto,gpu_memory_utilization=0.4,add_bos_token=True,max_model_len=4096 \ --tasks openllm \ --batch_size auto ``` ### Accuracy #### Open LLM Leaderboard evaluation scores
Benchmark SmolLM-1.7B-Instruct-quantized SmolLM-1.7B-Instruct-quantized.w8a16 (this model) Recovery
MMLU (5-shot) 28.10 28.42 101.1%
ARC Challenge (25-shot) 49.06 49.32 100.5%
GSM-8K (5-shot, strict-match) 4.93 4.93 100.0%
Hellaswag (10-shot) 66.96 66.89 99.9%
Winogrande (5-shot) 61.01 61.17 100.3%
TruthfulQA (0-shot) 40.28 40.25 99.4%
Average 41.76 41.83 100.2%