olmo3¶
OLMo 3 on vLLM (vllm.model_executor.models.olmo3).
Llama's names and a plain block: it is called with the positions and the residual stream and returns the stream. It is post-norm: each sublayer runs on the stream itself and what the block adds is the sublayer's norm's output, so the contributions point at the sibling norms, as on transformers. The attention norms its queries and keys.