MDSection 30 of 38 · 9 subtopics
Inference
Production generation: prefill, decode, batching, streaming, latency, and throughput.
Learning path
Work through each subtopic in order. Click any file below to open its lesson. Track your progress with the checkbox at the bottom.
Subtopics
Each subtopic includes a concise explanation, code examples, and references.
01
LLM Inference
Request-to-token pipeline.
llm-inference
02
Prefill
Parallel prompt-processing visualization.
prefill
03
Decode
Token-by-token generation visualization.
decode
04
Token Generation
Live token stream.
token-generation
05
Inference Latency
Latency timeline.
inference-latency
06
Throughput
Requests/tokens-per-second dashboard.
throughput
07
Batch Inference
Multiple-request batching animation.
batch-inference
08
Continuous Batching
Dynamic batch timeline.
continuous-batching
09
Streaming
Token-stream output.
streaming