Merge branch 'fix_tool_call'

tests/test_deepseek_ocr.rs

@@ -41,8 +41,14 @@ fn deepseek_ocr_generate() -> Result<()> {
     let i_start = Instant::now();
     let res = model.generate(mes)?;
     let i_duration = i_start.elapsed();
-    println!("Time elapsed in generate is: {:?}", i_duration);
     println!("generate: \n {:?}", res);
+    if res.usage.is_some() {
+        let num_token = res.usage.as_ref().unwrap().total_tokens;
+        let duration_secs = i_duration.as_secs_f64();
+        let tps = num_token as f64 / duration_secs;
+        println!("Tokens per second (TPS): {:.2}", tps);
+    }
+    println!("Time elapsed in generate is: {:?}", i_duration);    
     Ok(())
 }
 

tests/test_gelab_zero.rs

@@ -19,11 +19,39 @@ fn gelab_zero_generate() -> Result<()> {
                 "content": [    
                     {
                         "type": "text", 
-                        "text": "Hello, GELab-Zero!"
+                        "text": "Hello, GELab-Zero!, 现在几点了"
                     }
                 ]
             }
-        ]
+        ],
+        "tools": [
+            {
+                "type": "function",
+                "function": {
+                    "name": "get_current_time",
+                    "description": "当你想知道现在的时间时非常有用。",
+                    "parameters": {}
+                }
+            },
+            {
+                "type": "function",
+                "function": {
+                    "name": "get_current_weather",
+                    "description": "当你想查询指定城市的天气时非常有用。",
+                    "parameters": {
+                        "type": "object",
+                        "properties": {
+                            "location": {
+                                "type": "string",
+                                "description": "城市或县区，比如北京市、杭州市、余杭区等。"
+                            }
+                        },
+                        "required": ["location"]
+                    }
+                }
+            }
+        ],
+        "tool_choice": null
     }
     "#;
     let mes: ChatCompletionParameters = serde_json::from_str(message)?;
@@ -36,6 +64,12 @@ fn gelab_zero_generate() -> Result<()> {
     let res = qwen3vl.generate(mes)?;
     let i_duration = i_start.elapsed();
     println!("generate: \n {:?}", res);
+    if res.usage.is_some() {
+        let num_token = res.usage.as_ref().unwrap().total_tokens;
+        let duration_secs = i_duration.as_secs_f64();
+        let tps = num_token as f64 / duration_secs;
+        println!("Tokens per second (TPS): {:.2}", tps);
+    }
     println!("Time elapsed in generate is: {:?}", i_duration);
     Ok(())
 }

tests/test_hunyuan_ocr.rs

@@ -40,8 +40,15 @@ fn hunyuan_ocr_generate() -> Result<()> {
     let i_start = Instant::now();
     let res = model.generate(mes)?;
     let i_duration = i_start.elapsed();
-    println!("Time elapsed in generate is: {:?}", i_duration);
     println!("generate: \n {:?}", res);
+    if res.usage.is_some() {
+        let num_token = res.usage.as_ref().unwrap().total_tokens;
+        let duration_secs = i_duration.as_secs_f64();
+        let tps = num_token as f64 / duration_secs;
+        println!("Tokens per second (TPS): {:.2}", tps);
+    }
+    println!("Time elapsed in generate is: {:?}", i_duration);
+    
     Ok(())
 }
 

tests/test_minicpm4.rs

@@ -33,8 +33,14 @@ fn minicpm_generate() -> Result<()> {
 
     let i_start = Instant::now();
     let result = model.generate(mes)?;
-    println!("generate: \n {:?}", result);
     let i_duration = i_start.elapsed();
+    println!("generate: \n {:?}", result);
+    if result.usage.is_some() {
+        let num_token = result.usage.as_ref().unwrap().total_tokens;
+        let duration_secs = i_duration.as_secs_f64();
+        let tps = num_token as f64 / duration_secs;
+        println!("Tokens per second (TPS): {:.2}", tps);
+    }
     println!("Time elapsed in generate is: {:?}", i_duration);
 
     Ok(())

tests/test_paddleocr_vl.rs

@@ -41,8 +41,14 @@ fn paddleocr_vl_generate() -> Result<()> {
     let i_start = Instant::now();
     let res = model.generate(mes)?;
     let i_duration = i_start.elapsed();
-    println!("Time elapsed in generate is: {:?}", i_duration);
     println!("generate: \n {:?}", res);
+    if res.usage.is_some() {
+        let num_token = res.usage.as_ref().unwrap().total_tokens;
+        let duration_secs = i_duration.as_secs_f64();
+        let tps = num_token as f64 / duration_secs;
+        println!("Tokens per second (TPS): {:.2}", tps);
+    }
+    println!("Time elapsed in generate is: {:?}", i_duration);    
     Ok(())
 }
 

tests/test_qwen2_5vl.rs

@@ -45,9 +45,15 @@ fn qwen2_5vl_generate() -> Result<()> {
     println!("Time elapsed in load model is: {:?}", i_duration);
 
     let i_start = Instant::now();
-    let result = model.generate(mes)?;
-    println!("generate: \n {:?}", result);
+    let result = model.generate(mes)?;    
     let i_duration = i_start.elapsed();
+    println!("generate: \n {:?}", result);
+    if result.usage.is_some() {
+        let num_token = result.usage.as_ref().unwrap().total_tokens;
+        let duration_secs = i_duration.as_secs_f64();
+        let tps = num_token as f64 / duration_secs;
+        println!("Tokens per second (TPS): {:.2}", tps);
+    }
     println!("Time elapsed in generate is: {:?}", i_duration);
 
     Ok(())

tests/test_qwen3vl.rs

@@ -24,10 +24,10 @@ fn qwen3vl_generate() -> Result<()> {
                         {
                             "url": "./assets/video/video_test.mp4"
                         }
-                    },             
+                    },              
                     {
                         "type": "text", 
-                        "text": "视频里发生了什么"
+                        "text": "视频中发生了什么？, 现在几点了"
                     }
                 ]
             }
@@ -44,6 +44,12 @@ fn qwen3vl_generate() -> Result<()> {
     let res = qwen3vl.generate(mes)?;
     let i_duration = i_start.elapsed();
     println!("generate: \n {:?}", res);
+    if res.usage.is_some() {
+        let num_token = res.usage.as_ref().unwrap().total_tokens;
+        let duration_secs = i_duration.as_secs_f64();
+        let tps = num_token as f64 / duration_secs;
+        println!("Tokens per second (TPS): {:.2}", tps);
+    }
     println!("Time elapsed in generate is: {:?}", i_duration);
     Ok(())
 }
@@ -60,7 +66,7 @@ async fn qwen3vl_stream() -> Result<()> {
         "messages": [
             {
                 "role": "user",
-                "content": [ 
+                "content": [
                     {
                         "type": "video",
                         "video_url": 

tests/test_robo_brain.rs

@@ -34,8 +34,14 @@ fn robo_brain_generate() -> Result<()> {
 
     let i_start = Instant::now();
     let result = model.generate(mes)?;
-    println!("generate: \n {:?}", result);
     let i_duration = i_start.elapsed();
+    println!("generate: \n {:?}", result);    
+    if result.usage.is_some() {
+        let num_token = result.usage.as_ref().unwrap().total_tokens;
+        let duration_secs = i_duration.as_secs_f64();
+        let tps = num_token as f64 / duration_secs;
+        println!("Tokens per second (TPS): {:.2}", tps);
+    }
     println!("Time elapsed in generate is: {:?}", i_duration);
 
     Ok(())