For this study a random cohort of 206 German stroke CT reports was retrieved from local RIS/PACS (01/2015-12/2023). A stratified random split (90%-10%) using the ASPECT score [y|n], CTA/CTP [y|n] and sex[F|M] was performed to create training-validation (n=185) and test sets (n=21). The former was further stratified into training (n=160) and validation (n=25) sets. ICL-approaches were compared using 10 runs of a 0-shot- (validation fit/run only), 1-, 5-, and 10-shot disjunct training examples (n_runs=1+30;n_sum_examples=160). State-of-the-art OS-LLMs (n=8) including gemma2[:2b;9b], llama3.1[:8b;70b]|-3.2[3b], mistral[:7b;-small:22b] and mixtral[:8x7b] were tasked with summarising findings to the corresponding impressions. System-level role was provided in English and German (n=2). Models with top-3 validation performance were evaluated on the test set once. To ensure human interpretability, the cosine similarity index (CSI; range:0-1)was calculated instead of using comparisons in various LLM-embedding spaces.