Red Hat Summit이 콘텐츠는 선택한 언어로 제공되지 않습니다.
Chapter 10. Validated models for x86_64 CPU inference serving
The following large language models have been validated for use with Red Hat AI Inference Server on x86_64 CPUs with AVX2 instruction set support. CPU inference is optimized for smaller models that can run efficiently without GPU acceleration.
x86_64 CPU inference is best suited for smaller models, typically under 3 billion parameters. Performance depends on your CPU specifications, available system RAM, and model size. For larger models or production workloads requiring high throughput, consider using GPU acceleration.
{feature-name} is a Technology Preview feature only. Technology Preview features are not supported with Red Hat production service level agreements (SLAs) and might not be functionally complete. Red Hat does not recommend using them in production. These features provide early access to upcoming product features, enabling customers to test functionality and provide feedback during the development process.
For more information about the support scope of Red Hat Technology Preview features, see Technology Preview Features Support Scope.
| Model | Hugging Face model card | Number of parameters |
|---|---|---|
| TinyLlama-1.1B-Chat-v1.0 | 1.1B | |
| Llama-3.2-1B-Instruct | 1B | |
| granite-3.2-2b-instruct | 2B | |
| TinyLlama-1.1B-Chat-v1.0-pruned2.4 | 1.1B (pruned) | |
| TinyLlama-1.1B-Chat-v0.4-pruned50-quant-ds | 1.1B (pruned + quantized) | |
| opt-125m | 125M | |
| Qwen2-0.5B-Instruct-AWQ | 0.5B |
Quantization formats that require GPU-specific kernels, such as Marlin format, are not supported for CPU inference. Use AWQ or GPTQ quantization formats that are compatible with CPU execution.
The following table provides general guidance for approximate system RAM requirements based on model size:
| Model size | Minimum RAM | Recommended RAM |
|---|---|---|
| 125M - 500M | 8GB | 16GB |
| 500M - 1B | 16GB | 32GB |
| 1B - 3B | 32GB | 64GB |
Actual memory usage depends on the model architecture, context length, and batch size. Increase the VLLM_CPU_KVCACHE_SPACE environment variable to allocate more memory for the key-value cache when using longer context lengths.
'%20d='M201.5%20305.5c-13.8%200-24.9-11.1-24.9-24.6%200-13.8%2011.1-24.9%2024.9-24.9%2013.6%200%2024.6%2011.1%2024.6%2024.9%200%2013.6-11.1%2024.6-24.6%2024.6zM504%20256c0%20137-111%20248-248%20248S8%20393%208%20256%20119%208%20256%208s248%20111%20248%20248zm-132.3-41.2c-9.4%200-17.7%203.9-23.8%2010-22.4-15.5-52.6-25.5-86.1-26.6l17.4-78.3%2055.4%2012.5c0%2013.6%2011.1%2024.6%2024.6%2024.6%2013.8%200%2024.9-11.3%2024.9-24.9s-11.1-24.9-24.9-24.9c-9.7%200-18%205.8-22.1%2013.8l-61.2-13.6c-3-.8-6.1%201.4-6.9%204.4l-19.1%2086.4c-33.2%201.4-63.1%2011.3-85.5%2026.8-6.1-6.4-14.7-10.2-24.1-10.2-34.9%200-46.3%2046.9-14.4%2062.8-1.1%205-1.7%2010.2-1.7%2015.5%200%2052.6%2059.2%2095.2%20132%2095.2%2073.1%200%20132.3-42.6%20132.3-95.2%200-5.3-.6-10.8-1.9-15.8%2031.3-16%2019.8-62.5-14.9-62.5zM302.8%20331c-18.2%2018.2-76.1%2017.9-93.6%200-2.2-2.2-6.1-2.2-8.3%200-2.5%202.5-2.5%206.4%200%208.6%2022.8%2022.8%2087.3%2022.8%20110.2%200%202.5-2.2%202.5-6.1%200-8.6-2.2-2.2-6.1-2.2-8.3%200zm7.7-75c-13.6%200-24.6%2011.1-24.6%2024.9%200%2013.6%2011.1%2024.6%2024.6%2024.6%2013.8%200%2024.9-11.1%2024.9-24.6%200-13.8-11-24.9-24.9-24.9z'/%3e%3c/svg%3e){kind=small}