Offline Inference

Offline inference examples demonstrate how to use vLLM in an offline setting, where the model is queried for predictions in batches. We recommend starting with Basic.

Examples

• Audio Language
• Basic
• Batch LLM Inference
• Chat With Tools
• Data Parallel
• Disaggregated Prefill V1
• Disaggregated Prefill
• Eagle
• Embed Jina Embeddings V3
• Embed Matryoshka Fy
• Encoder Decoder
• Encoder Decoder Multimodal
• LLM Engine Example
• Load Sharded State
• LoRA With Quantization Inference
• Mistral-Small
• MLPSpeculator
• MultiLoRA Inference
• Neuron
• Neuron Eagle
• Neuron INT8 Quantization
• Neuron Speculation
• Offline Inference with the OpenAI Batch file format
• Prefix Caching
• Prithvi Geospatial MAE
• Profiling
• vLLM TPU Profiling
• Qwen2.5-Omni Offline Inference Examples
• Qwen 1M
• Reproducibility
• RLHF
• RLHF Colocate
• RLHF Utils
• Save Sharded State
• Simple Profiling
• Structured Outputs
• Torchrun Example
• TPU
• Vision Language
• Vision Language Embedding
• Vision Language Multi Image