in README-ollama.md: the httpie recipes to reach both servers

Author: parmentelat

notebooks/tps/chatbot/.teacher/README-ollama.md

@@ -20,3 +20,22 @@ ollama pull gemma2:2b
 ollama pull mistral:7b
 ollama pull deepseek-r1:7b
 ```
+
+## testing
+
+Using `http` (from `pip install httpie`) you can reach both instances from the command line like so:
+
+```bash
+
+# the CPU
+http http://ollama.pl.sophia.inria.fr:8080/api/generate model=mistral:7b prompt="Hey"
+
+# the GPU
+http https://Bob:[email protected]/api/generate model=mistral:7b prompt="Hey"
+
+# in streaming mode: CPU
+http -S http://ollama.pl.sophia.inria.fr:8080/api/generate model=mistral:7b prompt="Hey" stream:=true
+
+# in streaming mode: GPU
+http -S https://Bob:[email protected]/api/generate model=mistral:7b prompt="Hey" stream:=true
+```