fix aha run multiple inputs bug

docs/CLI_USAGE.md

@@ -65,7 +65,7 @@ aha -m qwen3vl-2b
 
 **语法：**
 ```bash
-aha run [OPTIONS] --model <MODEL> --input <INPUT> --weight-path <WEIGHT_PATH>
+aha run [OPTIONS] --model <MODEL> --input <INPUT> [--input <INPUT2>] --weight-path <WEIGHT_PATH>
 ```
 
 **选项：**
@@ -73,30 +73,39 @@ aha run [OPTIONS] --model <MODEL> --input <INPUT> --weight-path <WEIGHT_PATH>
 | 选项 | 说明 | 默认值 |
 |------|------|--------|
 | `-m, --model <MODEL>` | 模型类型（必选） | - |
-| `-in, --input <INPUT>` | 输入文本或文件路径（模型特定解释） | - |
+| `-i, --input <INPUT>` | 输入文本或文件路径（模型特定解释，支持1-2个参数, input1： 提示文本, input2: 文件地址） | - |
-| `-out, --output <OUTPUT>` | 输出文件路径（可选，未指定则自动生成） | - |
+| `-o, --output <OUTPUT>` | 输出文件路径（可选，未指定则自动生成） | - |
 | `--weight-path <WEIGHT_PATH>` | 本地模型权重路径（必选） | - |
 
 **示例：**
 
 ```bash
-# VoxCPM1.5 文字转语音
+# VoxCPM1.5 文字转语音（单个输入）
-aha run -m voxcpm1.5 -in "太阳当空照" -out output.wav --weight-path /path/to/model
+aha run -m voxcpm1.5 -i "太阳当空照" -o output.wav --weight-path /path/to/model
 
-# VoxCPM1.5 从文件读取输入
+# VoxCPM1.5 从文件读取输入（单个输入）
-aha run -m voxcpm1.5 -in "file://./input.txt" --weight-path /path/to/model
+aha run -m voxcpm1.5 -i "file://./input.txt" --weight-path /path/to/model
 
-# MiniCPM4 文本生成
+# MiniCPM4 文本生成（单个输入）
-aha run -m minicpm4-0.5b -in "你好" --weight-path /path/to/model
+aha run -m minicpm4-0.5b -i "你好" --weight-path /path/to/model
 
-# DeepSeek OCR 图片识别
+# DeepSeek OCR 图片识别（单个输入）
-aha run -m deepseek-ocr -in "image.jpg" --weight-path /path/to/model
+aha run -m deepseek-ocr -i "image.jpg" --weight-path /path/to/model
 
-# RMBG2.0 背景移除
+# RMBG2.0 背景移除（单个输入）
-aha run -m RMBG2.0 -in "photo.png" -out "no_bg.png" --weight-path /path/to/model
+aha run -m RMBG2.0 -i "photo.png" -o "no_bg.png" --weight-path /path/to/model
 
-# GLM-ASR 语音识别
+# GLM-ASR 语音识别（两个输入：提示文本 + 音频文件）
-aha run -m glm-asr-nano-2512 -in "audio.wav" -in "请转写这段音频" --weight-path /path/to/model
+aha run -m glm-asr-nano-2512 -i "请转写这段音频" -i "audio.wav" --weight-path /path/to/model
+
+# Fun-ASR 语音识别（两个输入：提示文本 + 音频文件）
+aha run -m fun-asr-nano-2512 -i "语音转写：" -i "audio.wav" --weight-path /path/to/model
+
+# qwen3 文本生成（单个输入）
+aha run -m qwen3-0.6b -i "你好" --weight-path /path/to/model
+
+# qwen2.5vl 图像理解（两个输入：提示文本 + 图片文件）
+aha run -m qwen2.5vl-3b -i "请分析图片并提取所有可见文本内容，按从左到右、从上到下的布局，返回纯文本" -i "image.jpg"  --weight-path /path/to/model
 ```
 
 ### serv - 启动服务

src/exec/deepseek_ocr.rs

@@ -1,18 +1,24 @@
 //! DeepSeek-OCR exec implementation for CLI `run` subcommand
 
+use std::time::Instant;
+
+use anyhow::{Ok, Result};
+
 use crate::exec::ExecModel;
 use crate::models::{GenerateModel, deepseek_ocr::generate::DeepseekOCRGenerateModel};
-use anyhow::{Ok, Result};
-use std::time::Instant;
 
 pub struct DeepSeekORExec;
 
 impl ExecModel for DeepSeekORExec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let input_path = if input.starts_with("file://") {
+        let url = &input[0];
-            input.to_string()
+        let input_url = if url.starts_with("http://")
+            || url.starts_with("https://")
+            || url.starts_with("file://")
+        {
+            url.clone()
         } else {
-            format!("file://{}", input)
+            format!("file://{}", url)
         };
 
         let i_start = Instant::now();
@@ -32,12 +38,16 @@ impl ExecModel for DeepSeekORExec {
                             "image_url": {{
                                 "url": "{}"
                             }}
+                        }},
+                        {{
+                            "type": "text", 
+                            "text": "<image>\nConvert the document to markdown. "
                         }}
                     ]
                 }}
             ]
         }}"#,
-            input_path
+            input_url
         );
         let mes = serde_json::from_str(&message)?;
 

src/exec/fun_asr_nano.rs

@@ -1,19 +1,24 @@
 //! Fun-ASR-Nano-2512 exec implementation for CLI `run` subcommand
 
+use std::time::Instant;
+
+use anyhow::{Ok, Result};
+
 use crate::exec::ExecModel;
 use crate::models::{GenerateModel, fun_asr_nano::generate::FunAsrNanoGenerateModel};
-use anyhow::{Ok, Result};
+use crate::utils::get_file_path;
-use std::time::Instant;
 
 pub struct FunASRNanoExec;
 
 impl ExecModel for FunASRNanoExec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let target_text = if input.starts_with("file://") {
+        let input_text = &input[0];
-            let path = &input[7..];
+        let target_text = if input_text.starts_with("file://") {
+            // let path = &input[7..];
+            let path = get_file_path(input_text)?;
             std::fs::read_to_string(path)?
         } else {
-            input.to_string()
+            input_text.clone()
         };
 
         let i_start = Instant::now();
@@ -22,10 +27,14 @@ impl ExecModel for FunASRNanoExec {
         println!("Time elapsed in load model is: {:?}", i_duration);
 
         // Create ChatCompletionParameters for ASR
-        let input_url = if input.starts_with("http://") || input.starts_with("https://") || input.starts_with("file://") {
+        let url = &input[1];
-            input.to_string()
+        let input_url = if url.starts_with("http://")
+            || url.starts_with("https://")
+            || url.starts_with("file://")
+        {
+            url.clone()
         } else {
-            format!("file://{}", input)
+            format!("file://{}", url)
         };
 
         let message = format!(

src/exec/glm_asr_nano.rs

@@ -1,19 +1,24 @@
 //! GLM-ASR-Nano-2512 exec implementation for CLI `run` subcommand
 
+use std::time::Instant;
+
+use anyhow::{Ok, Result};
+
 use crate::exec::ExecModel;
 use crate::models::{GenerateModel, glm_asr_nano::generate::GlmAsrNanoGenerateModel};
-use anyhow::{Ok, Result};
+use crate::utils::get_file_path;
-use std::time::Instant;
 
 pub struct GlmASRNanoExec;
 
 impl ExecModel for GlmASRNanoExec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let target_text = if input.starts_with("file://") {
+        let input_text = &input[0];
-            let path = &input[7..];
+        let target_text = if input_text.starts_with("file://") {
+            // let path = &input[7..];
+            let path = get_file_path(input_text)?;
             std::fs::read_to_string(path)?
         } else {
-            input.to_string()
+            input_text.clone()
         };
 
         let i_start = Instant::now();
@@ -23,10 +28,14 @@ impl ExecModel for GlmASRNanoExec {
 
         // Create ChatCompletionParameters for ASR
         // Input should be an audio file path
-        let input_url = if input.starts_with("http://") || input.starts_with("https://") || input.starts_with("file://") {
+        let url = &input[1];
-            input.to_string()
+        let input_url = if url.starts_with("http://")
+            || url.starts_with("https://")
+            || url.starts_with("file://")
+        {
+            url.clone()
         } else {
-            format!("file://{}", input)
+            format!("file://{}", url)
         };
 
         let message = format!(

src/exec/hunyuan_ocr.rs

@@ -1,18 +1,24 @@
 //! Hunyuan-OCR exec implementation for CLI `run` subcommand
 
+use std::time::Instant;
+
+use anyhow::{Ok, Result};
+
 use crate::exec::ExecModel;
 use crate::models::{GenerateModel, hunyuan_ocr::generate::HunyuanOCRGenerateModel};
-use anyhow::{Ok, Result};
-use std::time::Instant;
 
 pub struct HunyuanORExec;
 
 impl ExecModel for HunyuanORExec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let input_path = if input.starts_with("file://") {
+        let url = &input[0];
-            input.to_string()
+        let input_url = if url.starts_with("http://")
+            || url.starts_with("https://")
+            || url.starts_with("file://")
+        {
+            url.clone()
         } else {
-            format!("file://{}", input)
+            format!("file://{}", url)
         };
 
         let i_start = Instant::now();
@@ -32,12 +38,16 @@ impl ExecModel for HunyuanORExec {
                             "image_url": {{
                                 "url": "{}"
                             }}
+                        }},
+                        {{
+                            "type": "text", 
+                            "text": "检测并识别图片中的文字，将文本坐标格式化输出。"
                         }}
                     ]
                 }}
             ]
         }}"#,
-            input_path
+            input_url
         );
         let mes = serde_json::from_str(&message)?;
 

src/exec/minicpm4.rs

@@ -1,19 +1,24 @@
 //! MiniCPM4-0.5B exec implementation for CLI `run` subcommand
 
+use std::time::Instant;
+
+use anyhow::{Ok, Result};
+
 use crate::exec::ExecModel;
 use crate::models::{GenerateModel, minicpm4::generate::MiniCPMGenerateModel};
-use anyhow::{Ok, Result};
+use crate::utils::get_file_path;
-use std::time::Instant;
 
 pub struct MiniCPM4Exec;
 
 impl ExecModel for MiniCPM4Exec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let target_text = if input.starts_with("file://") {
+        let input_text = &input[0];
-            let path = &input[7..];
+        let target_text = if input_text.starts_with("file://") {
+            // let path = &input[7..];
+            let path = get_file_path(input_text)?;
             std::fs::read_to_string(path)?
         } else {
-            input.to_string()
+            input_text.to_string()
         };
 
         let i_start = Instant::now();

src/exec/mod.rs

@@ -33,5 +33,5 @@ pub trait ExecModel {
     /// # Returns
     /// * `Ok(())` on success
     /// * `Err(anyhow::Error)` on failure
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()>;
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()>;
 }

src/exec/paddleocr_vl.rs

@@ -1,18 +1,24 @@
 //! PaddleOCR-VL exec implementation for CLI `run` subcommand
 
+use std::time::Instant;
+
+use anyhow::{Ok, Result};
+
 use crate::exec::ExecModel;
 use crate::models::{GenerateModel, paddleocr_vl::generate::PaddleOCRVLGenerateModel};
-use anyhow::{Ok, Result};
-use std::time::Instant;
 
 pub struct PaddleOVLExec;
 
 impl ExecModel for PaddleOVLExec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let input_path = if input.starts_with("file://") {
+        let url = &input[0];
-            input.to_string()
+        let input_url = if url.starts_with("http://")
+            || url.starts_with("https://")
+            || url.starts_with("file://")
+        {
+            url.clone()
         } else {
-            format!("file://{}", input)
+            format!("file://{}", url)
         };
 
         let i_start = Instant::now();
@@ -32,12 +38,16 @@ impl ExecModel for PaddleOVLExec {
                             "image_url": {{
                                 "url": "{}"
                             }}
+                        }},
+                        {{
+                            "type": "text", 
+                            "text": "OCR:"
                         }}
                     ]
                 }}
             ]
         }}"#,
-            input_path
+            input_url
         );
         let mes = serde_json::from_str(&message)?;
 

src/exec/qwen2_5vl.rs

@@ -1,21 +1,33 @@
 //! Qwen2.5VL-3B exec implementation for CLI `run` subcommand
 
+use std::time::Instant;
+
+use anyhow::{Ok, Result};
+
 use crate::exec::ExecModel;
 use crate::models::{GenerateModel, qwen2_5vl::generate::Qwen2_5VLGenerateModel};
-use anyhow::{Ok, Result};
+use crate::utils::get_file_path;
-use std::time::Instant;
 
 pub struct Qwen2_5vlExec;
 
 impl ExecModel for Qwen2_5vlExec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let target_text = if input.starts_with("file://") {
+        let input_text = &input[0];
-            let path = &input[7..];
+        let target_text = if input_text.starts_with("file://") {
+            let path = get_file_path(input_text)?;
             std::fs::read_to_string(path)?
         } else {
-            input.to_string()
+            input_text.clone()
+        };
+        let url = &input[1];
+        let input_url = if url.starts_with("http://")
+            || url.starts_with("https://")
+            || url.starts_with("file://")
+        {
+            url.clone()
+        } else {
+            format!("file://{}", url)
         };
-
         let i_start = Instant::now();
         let mut model = Qwen2_5VLGenerateModel::init(weight_path, None, None)?;
         let i_duration = i_start.elapsed();
@@ -27,11 +39,22 @@ impl ExecModel for Qwen2_5vlExec {
             "messages": [
                 {{
                     "role": "user",
-                    "content": "{}"
+                    "content": [
+                        {{
+                            "type": "image",
+                            "image_url": {{
+                                "url": "{}"
+                            }}
+                        }},
+                        {{
+                            "type": "text", 
+                            "text": "{}"
+                        }}
+                    ]
                 }}
             ]
         }}"#,
-            target_text.replace('"', "\\\"")
+            input_url, target_text
         );
         let mes = serde_json::from_str(&message)?;
 

src/exec/qwen3.rs

@@ -1,19 +1,24 @@
 //! Qwen3-0.6B exec implementation for CLI `run` subcommand
 
+use std::time::Instant;
+
+use anyhow::{Ok, Result};
+
 use crate::exec::ExecModel;
 use crate::models::{GenerateModel, qwen3::generate::Qwen3GenerateModel};
-use anyhow::{Ok, Result};
+use crate::utils::get_file_path;
-use std::time::Instant;
 
 pub struct Qwen3Exec;
 
 impl ExecModel for Qwen3Exec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let target_text = if input.starts_with("file://") {
+        let input_text = &input[0];
-            let path = &input[7..];
+        let target_text = if input_text.starts_with("file://") {
+            // let path = &input[7..];
+            let path = get_file_path(input_text)?;
             std::fs::read_to_string(path)?
         } else {
-            input.to_string()
+            input_text.clone()
         };
 
         let i_start = Instant::now();

src/exec/qwen3vl.rs

@@ -1,38 +1,88 @@
 //! Qwen3VL-2B exec implementation for CLI `run` subcommand
 
+use std::time::Instant;
+
+use anyhow::{Ok, Result};
+
 use crate::exec::ExecModel;
 use crate::models::{GenerateModel, qwen3vl::generate::Qwen3VLGenerateModel};
-use anyhow::{Ok, Result};
+use crate::utils::get_file_path;
-use std::time::Instant;
 
 pub struct Qwen3vlExec;
 
 impl ExecModel for Qwen3vlExec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let target_text = if input.starts_with("file://") {
+        let input_text = &input[0];
-            let path = &input[7..];
+        let target_text = if input_text.starts_with("file://") {
+            let path = get_file_path(input_text)?;
             std::fs::read_to_string(path)?
         } else {
-            input.to_string()
+            input_text.clone()
         };
 
         let i_start = Instant::now();
         let mut model = Qwen3VLGenerateModel::init(weight_path, None, None)?;
         let i_duration = i_start.elapsed();
         println!("Time elapsed in load model is: {:?}", i_duration);
-
+        let url = &input[1];
-        let message = format!(
+        let input_url = if url.starts_with("http://")
-            r#"{{
+            || url.starts_with("https://")
+            || url.starts_with("file://")
+        {
+            url.clone()
+        } else {
+            format!("file://{}", url)
+        };
+        let message = if input_url.ends_with("mp4") {
+            format!(
+                r#"{{
             "model": "qwen3vl",
             "messages": [
                 {{
                     "role": "user",
-                    "content": "{}"
+                    "content": [
+                        {{
+                            "type": "video",
+                            "video_url": 
+                            {{
+                                "url": "{}"
+                            }}
+                        }},
+                        {{
+                            "type": "text", 
+                            "text": "{}"
+                        }}
+                    ]
                 }}
             ]
         }}"#,
-            target_text.replace('"', "\\\"")
+                input_url, target_text
-        );
+            )
+        } else {
+            format!(
+                r#"{{
+            "model": "qwen2.5vl",
+            "messages": [
+                {{
+                    "role": "user",
+                    "content": [
+                        {{
+                            "type": "image",
+                            "image_url": {{
+                                "url": "{}"
+                            }}
+                        }},
+                        {{
+                            "type": "text", 
+                            "text": "{}"
+                        }}
+                    ]
+                }}
+            ]
+        }}"#,
+                input_url, target_text
+            )
+        };
         let mes = serde_json::from_str(&message)?;
 
         let i_start = Instant::now();

src/exec/rmbg2_0.rs

@@ -1,18 +1,24 @@
 //! RMBG2.0 exec implementation for CLI `run` subcommand
 
+use std::time::Instant;
+
+use anyhow::{Ok, Result};
+
 use crate::exec::ExecModel;
 use crate::models::rmbg2_0::generate::RMBG2_0Model;
-use anyhow::{Ok, Result};
-use std::time::Instant;
 
 pub struct RMBG2_0Exec;
 
 impl ExecModel for RMBG2_0Exec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let input_path = if input.starts_with("file://") {
+        let url = &input[0];
-            input.to_string()
+        let input_url = if url.starts_with("http://")
+            || url.starts_with("https://")
+            || url.starts_with("file://")
+        {
+            url.clone()
         } else {
-            format!("file://{}", input)
+            format!("file://{}", url)
         };
 
         let i_start = Instant::now();
@@ -38,7 +44,7 @@ impl ExecModel for RMBG2_0Exec {
                 }}
             ]
         }}"#,
-            input_path
+            input_url
         );
         let mes = serde_json::from_str(&message)?;
 

src/exec/voxcpm.rs

@@ -1,19 +1,23 @@
 //! VoxCPM exec implementation for CLI `run` subcommand
 
-use crate::exec::ExecModel;
-use crate::models::voxcpm::generate::VoxCPMGenerate;
-use anyhow::{Ok, Result};
 use std::time::Instant;
 
+use anyhow::{Ok, Result};
+
+use crate::models::voxcpm::generate::VoxCPMGenerate;
+use crate::{exec::ExecModel, utils::get_file_path};
+
 pub struct VoxCPMExec;
 
 impl ExecModel for VoxCPMExec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let target_text = if input.starts_with("file://") {
+        let input_text = &input[0];
-            let path = &input[7..];
+        let target_text = if input_text.starts_with("file://") {
+            // let path = &input[7..];
+            let path = get_file_path(input_text)?;
             std::fs::read_to_string(path)?
         } else {
-            input.to_string()
+            input_text.clone()
         };
 
         let i_start = Instant::now();
@@ -24,10 +28,10 @@ impl ExecModel for VoxCPMExec {
         let i_start = Instant::now();
         let audio = voxcpm_generate.inference(
             target_text,
-            Some("啥子小师叔，打狗还要看主人，你再要继续，我就是你的对手".to_string()), //todo args
+            Some("啥子小师叔，打狗还要看主人，你再要继续，我就是你的对手".to_string()), // todo args
-            Some("file://./assets/audio/voice_01.wav".to_string()),  //todo args
+            Some("file://./assets/audio/voice_01.wav".to_string()),                     // todo args
             2,
-            100,   // max_len (voxcpm uses 100 vs voxcpm1.5's 4096)
+            100, // max_len (voxcpm uses 100 vs voxcpm1.5's 4096)
             10,
             2.0,
             6.0,

src/exec/voxcpm1_5.rs

@@ -5,20 +5,24 @@
 //! - Input can be text content or a file path (with `file://` prefix)
 //! - Output can be a file path or will be auto-generated if not specified
 
-use crate::exec::ExecModel;
-use crate::models::voxcpm::generate::VoxCPMGenerate;
-use anyhow::{Ok, Result};
 use std::time::Instant;
 
+use anyhow::{Ok, Result};
+
+use crate::models::voxcpm::generate::VoxCPMGenerate;
+use crate::{exec::ExecModel, utils::get_file_path};
+
 pub struct VoxCPM1_5Exec;
 
 impl ExecModel for VoxCPM1_5Exec {
-    fn run(input: &str, output: Option<&str>, weight_path: &str) -> Result<()> {
+    fn run(input: &[String], output: Option<&str>, weight_path: &str) -> Result<()> {
-        let target_text = if input.starts_with("file://") {
+        let input_text = &input[0];
-            let path = &input[7..];
+        let target_text = if input_text.starts_with("file://") {
+            // let path = &input[7..];
+            let path = get_file_path(input_text)?;
             std::fs::read_to_string(path)?
         } else {
-            input.to_string()
+            input_text.clone()
         };
 
         let i_start = Instant::now();
@@ -29,8 +33,8 @@ impl ExecModel for VoxCPM1_5Exec {
         let i_start = Instant::now();
         let audio = voxcpm_generate.inference(
             target_text,
-            Some("啥子小师叔，打狗还要看主人，你再要继续，我就是你的对手".to_string()), //todo args
+            Some("啥子小师叔，打狗还要看主人，你再要继续，我就是你的对手".to_string()), // todo args
-            Some("file://./assets/audio/voice_01.wav".to_string()),  //todo args
+            Some("file://./assets/audio/voice_01.wav".to_string()),                     // todo args
             2,
             4096,
             10,

src/main.rs

@@ -129,8 +129,8 @@ struct RunArgs {
     model: WhichModel,
 
     /// Input text or file path
-    #[arg(short, long)]
+    #[arg(short, long, num_args = 1..=2, value_delimiter = ' ')]
-    input: String,
+    input: Vec<String>,
 
     /// Output file path (optional)
     #[arg(short, long)]
@@ -219,7 +219,7 @@ fn run_list() -> anyhow::Result<()> {
 
     println!("Available models:");
     println!();
-    println!("{:<30} {}", "Model Name", "ModelScope ID");
+    println!("{:<30} ModelScope ID", "Model Name");
     println!("{}", "-".repeat(80));
     for model in models {
         let possible_value = model.to_possible_value().unwrap();
@@ -233,7 +233,12 @@ fn run_list() -> anyhow::Result<()> {
 
 /// Run the 'cli' subcommand: download model (if needed) and start service
 async fn run_cli(args: CliArgs) -> anyhow::Result<()> {
-    let CliArgs { common, weight_path, save_dir, download_retries } = args;
+    let CliArgs {
+        common,
+        weight_path,
+        save_dir,
+        download_retries,
+    } = args;
     let model_id = get_model_id(common.model);
 
     let model_path = match weight_path {
@@ -257,7 +262,10 @@ async fn run_cli(args: CliArgs) -> anyhow::Result<()> {
 
 /// Run the 'serv' subcommand: start service only (no download)
 async fn run_serv(args: ServArgs) -> anyhow::Result<()> {
-    let ServArgs { common, weight_path } = args;
+    let ServArgs {
+        common,
+        weight_path,
+    } = args;
 
     init(common.model, weight_path)?;
     start_http_server(common.address, common.port).await?;
@@ -267,7 +275,11 @@ async fn run_serv(args: ServArgs) -> anyhow::Result<()> {
 
 /// Run the 'download' subcommand: download model only (no server)
 async fn run_download(args: DownloadArgs) -> anyhow::Result<()> {
-    let DownloadArgs { model, save_dir, download_retries } = args;
+    let DownloadArgs {
+        model,
+        save_dir,
+        download_retries,
+    } = args;
     let model_id = get_model_id(model);
 
     let save_dir = match save_dir {
@@ -285,7 +297,12 @@ async fn run_download(args: DownloadArgs) -> anyhow::Result<()> {
 fn run_run(args: RunArgs) -> anyhow::Result<()> {
     use aha::exec::ExecModel;
 
-    let RunArgs { model, input, output, weight_path } = args;
+    let RunArgs {
+        model,
+        input,
+        output,
+        weight_path,
+    } = args;
 
     match model {
         WhichModel::MiniCPM4_0_5B => {
@@ -405,4 +422,4 @@ pub(crate) async fn start_http_server(address: String, port: u16) -> anyhow::Res
 
     builder.launch().await?;
     Ok(())
-}
+}

src/utils/mod.rs

@@ -3,7 +3,7 @@ pub mod img_utils;
 pub mod tensor_utils;
 pub mod video_utils;
 
-use std::{fs, process::Command};
+use std::{fs, path::PathBuf, process::Command};
 
 use aha_openai_dive::v1::resources::{
     chat::{
@@ -490,3 +490,17 @@ pub fn get_default_save_dir() -> Option<String> {
         path.to_string_lossy().to_string()
     })
 }
+
+pub fn get_file_path(file: &str) -> Result<PathBuf> {
+    let path = url::Url::parse(file)?;
+    let path = path.to_file_path();
+    let path = match path {
+        Ok(path) => path,
+        Err(_) => {
+            let mut path = file.to_owned();
+            path = path.split_off(7);
+            PathBuf::from(path)
+        }
+    };
+    Ok(path)
+}