update dependences

src/models/minicpm4/generate.rs

@@ -1,9 +1,9 @@
+use aha_openai_dive::v1::resources::chat::{
+    ChatCompletionChunkResponse, ChatCompletionParameters, ChatCompletionResponse,
+};
 use anyhow::{Result, anyhow};
 use candle_core::{DType, Device, Tensor};
 use candle_nn::VarBuilder;
-use openai_dive::v1::resources::chat::{
-    ChatCompletionChunkResponse, ChatCompletionParameters, ChatCompletionResponse,
-};
 use rocket::async_stream::stream;
 use rocket::futures::Stream;
 

src/models/mod.rs

@@ -4,10 +4,10 @@ pub mod qwen2_5vl;
 pub mod qwen3vl;
 pub mod voxcpm;
 
-use anyhow::Result;
-use openai_dive::v1::resources::chat::{
+use aha_openai_dive::v1::resources::chat::{
     ChatCompletionChunkResponse, ChatCompletionParameters, ChatCompletionResponse,
 };
+use anyhow::Result;
 use rocket::futures::Stream;
 
 pub trait GenerateModel {

src/models/qwen2_5vl/generate.rs

@@ -1,10 +1,10 @@
 // use crate::models::GenerateStream;
+use aha_openai_dive::v1::resources::chat::{
+    ChatCompletionChunkResponse, ChatCompletionParameters, ChatCompletionResponse,
+};
 use anyhow::{Result, anyhow};
 use candle_core::{D, DType, Device, IndexOp, Tensor};
 use candle_nn::VarBuilder;
-use openai_dive::v1::resources::chat::{
-    ChatCompletionChunkResponse, ChatCompletionParameters, ChatCompletionResponse,
-};
 use rocket::async_stream::stream;
 use rocket::futures::Stream;
 

src/models/qwen2_5vl/processor.rs

@@ -1,13 +1,13 @@
 use std::collections::HashMap;
 
+use aha_openai_dive::v1::resources::chat::{
+    ChatCompletionParameters, ChatMessage, ChatMessageContent, ChatMessageContentPart,
+};
 use anyhow::{Result, anyhow};
 use candle_core::{DType, Device, IndexOp, Shape, Tensor};
 use ffmpeg_next as ffmpeg;
 use image::DynamicImage;
 use num::integer::lcm;
-use openai_dive::v1::resources::chat::{
-    ChatCompletionParameters, ChatMessage, ChatMessageContent, ChatMessageContentPart,
-};
 
 use crate::{
     models::qwen2_5vl::config::VisionSetting,

src/models/qwen3vl/generate.rs

@@ -1,9 +1,9 @@
+use aha_openai_dive::v1::resources::chat::{
+    ChatCompletionChunkResponse, ChatCompletionParameters, ChatCompletionResponse,
+};
 use anyhow::{Result, anyhow};
 use candle_core::{DType, Device, Tensor};
 use candle_nn::VarBuilder;
-use openai_dive::v1::resources::chat::{
-    ChatCompletionChunkResponse, ChatCompletionParameters, ChatCompletionResponse,
-};
 use rocket::async_stream::stream;
 use rocket::futures::Stream;
 

src/models/qwen3vl/processor.rs

@@ -1,13 +1,13 @@
 use std::collections::HashMap;
 
+use aha_openai_dive::v1::resources::chat::{
+    ChatCompletionParameters, ChatMessage, ChatMessageContent, ChatMessageContentPart,
+};
 use anyhow::{Result, anyhow};
 use candle_core::{DType, Device, IndexOp, Shape, Tensor};
 use ffmpeg_next as ffmpeg;
 use image::DynamicImage;
 use num::integer::lcm;
-use openai_dive::v1::resources::chat::{
-    ChatCompletionParameters, ChatMessage, ChatMessageContent, ChatMessageContentPart,
-};
 
 use crate::{
     models::qwen3vl::config::PreprocessorConfig,