Fix: Upstream URL for llama2 completion example (#7242)

* Update _llama2.md The example provided in the curl command references the ollama endpoint, http://ollama-server.local:11434/v1/chat, which returns a 404. Per the docs (referenced earlier on this page), the correct endpoint should be /api/chat which I can confirm works. https://github.com/ollama/ollama/blob/main/docs/api.md#generate-a-chat-completion * fix upstream url in curl example --------- Co-authored-by: lena-larionova <[email protected]>
Kong · Apr 16, 2024 · 0434574 · 0434574
1 parent aec7e38
commit 0434574
Showing 1 changed file with 3 additions and 3 deletions.
diff --git a/app/_hub/kong-inc/ai-proxy/how-to/_llama2.md b/app/_hub/kong-inc/ai-proxy/how-to/_llama2.md
@@ -89,7 +89,7 @@ curl -X POST http://localhost:8001/routes/llama2-chat/plugins \
   --data "config.model.provider=llama2" \
   --data "config.model.name=llama2" \
   --data "config.model.options.llama2_format=ollama" \
-  --data "config.model.options.upstream_url=http://ollama-server.local:11434/v1/chat" \ 
+  --data "config.model.options.upstream_url=http://ollama-server.local:11434/api/chat" \ 
 ```
 
 {% endnavtab %}
@@ -108,7 +108,7 @@ plugins:
         provider: "llama2"
         name: "llama2"
         llama2_format: "ollama"
-        upstream_url: "http://llama2-server.local:11434/v1/chat"
+        upstream_url: "http://llama2-server.local:11434/api/chat"
 ```
 {% endnavtab %}
 {% endnavtabs %}
@@ -121,4 +121,4 @@ Make an `llm/v1/chat` type request to test your new endpoint:
 curl -X POST http://localhost:8000/llama2-chat \
   -H 'Content-Type: application/json' \
   --data-raw '{ "messages": [ { "role": "system", "content": "You are a mathematician" }, { "role": "user", "content": "What is 1+1?"} ] }'
-```
+```