more changes

crates/Cargo.lock

@@ -1076,7 +1076,6 @@ checksum = "95505c38b4572b2d910cecb0281560f54b440a19336cbbcb27bf6ce6adc6f5a8"
 name = "hermesllm"
 version = "0.1.0"
 dependencies = [
- "log",
  "serde",
  "serde_json",
  "serde_with",

crates/hermesllm/Cargo.toml

@@ -4,7 +4,6 @@ version = "0.1.0"
 edition = "2021"
 
 [dependencies]
-log = "0.4.27"
 serde = "1.0.219"
 serde_json = "1.0.140"
 serde_with = "3.12.0"

crates/llm_gateway/src/stream_context.rs

@@ -599,7 +599,8 @@ impl HttpContext for StreamContext {
                     Err(e) => {
                         warn!("could not parse response: {}", e);
                         debug!(
-                            "on_http_response_body: response body: {}",
+                            "on_http_response_body: S[{}], response body: {}",
+                            self.context_id,
                             String::from_utf8_lossy(&body)
                         );
                         self.send_server_error(

tests/rest/api_llm_gateway.rest

@@ -2,8 +2,33 @@
 @openai_endpoint = https://api.openai.com
 @access_key = {{$dotenv OPENAI_API_KEY}}
 
-### openai request
-POST {{openai_endpoint}}/v1/chat/completions HTTP/1.1
+POST {{llm_endpoint}}/v1/chat/completions HTTP/1.1
+content-type: application/json
+authorization: Bearer
+accept: */*
+accept-encoding: deflate
+user-agent: Python/3.11 aiohttp/3.11.11
+content-length: 876
+x-forwarded-proto: https
+x-request-id: 99d7817d-a646-9497-a38d-710b1ce1325f
+traceparent: 00-e4c9fc8cf9fc7714c6a15ef34852fb30-573a351a98e0cd01-01
+tracestate:
+x-arch-llm-provider-hint: gpt-4o-mini
+
+
+{
+  "model": "gpt-4o-mini",
+  "messages": [
+    {
+      "role": "user",
+      "content": "### Task:\nGenerate 1-3 broad tags categorizing the main themes of the chat history, along with 1-3 more specific subtopic tags.\n\n### Guidelines:\n- Start with high-level domains (e.g. Science, Technology, Philosophy, Arts, Politics, Business, Health, Sports, Entertainment, Education)\n- Consider including relevant subfields/subdomains if they are strongly represented throughout the conversation\n- If content is too short (less than 3 messages) or too diverse, use only [\"General\"]\n- Use the chat's primary language; default to English if multilingual\n- Prioritize accuracy over specificity\n\n### Output:\nJSON format: { \"tags\": [\"tag1\", \"tag2\", \"tag3\"] }\n\n### Chat History:\n<chat_history>\nUSER: hello\nASSISTANT: Hello! How can I assist you today?\n</chat_history>"
+    }
+  ],
+  "stream": false
+}
+
+### test
+POST {{llm_endpoint}}/v1/chat/completions HTTP/1.1
 Content-Type: application/json
 Authorization: Bearer {{access_key}}
 
@@ -15,7 +40,7 @@ Authorization: Bearer {{access_key}}
     }
   ],
   "model": "gpt-4o-mini",
-  "stream": true
+  "stream": false
 }
 
 ### openai request (streaming)