update dependences

Cargo.lock

@@ -21,6 +21,7 @@ checksum = "320119579fcad9c21884f5c4861d16174d0e06250625266f50fe6898340abefa"
 name = "aha"
 version = "0.1.1"
 dependencies = [
+ "aha_openai_dive",
  "anyhow",
  "base64 0.22.1",
  "candle-core",
@@ -33,7 +34,6 @@ dependencies = [
  "image",
  "minijinja",
  "num",
- "openai_dive",
  "reqwest",
  "rocket",
  "serde",
@@ -43,6 +43,25 @@ dependencies = [
  "uuid",
 ]
 
+[[package]]
+name = "aha_openai_dive"
+version = "1.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "45b54fac70f40807364efeb885a33150e163d86b9db687820ba3c5f06aeeaf6e"
+dependencies = [
+ "bytes",
+ "derive_builder",
+ "futures",
+ "reqwest",
+ "reqwest-eventsource",
+ "serde",
+ "serde_html_form",
+ "serde_json",
+ "tokio",
+ "tokio-stream",
+ "tokio-util",
+]
+
 [[package]]
 name = "ahash"
 version = "0.8.12"
@@ -843,7 +862,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "39cab71617ae0d63f51a36d69f866391735b51691dbda63cf6f96d042b63efeb"
 dependencies = [
  "libc",
- "windows-sys 0.59.0",
+ "windows-sys 0.61.2",
 ]
 
 [[package]]
@@ -1643,7 +1662,7 @@ dependencies = [
  "libc",
  "percent-encoding",
  "pin-project-lite",
- "socket2 0.5.10",
+ "socket2 0.6.1",
  "system-configuration",
  "tokio",
  "tower-service",
@@ -1894,7 +1913,7 @@ checksum = "3640c1c38b8e4e43584d8df18be5fc6b0aa314ce6ebf51b53313d4306cca8e46"
 dependencies = [
  "hermit-abi",
  "libc",
- "windows-sys 0.59.0",
+ "windows-sys 0.61.2",
 ]
 
 [[package]]
@@ -2253,7 +2272,7 @@ version = "0.50.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "7957b9740744892f114936ab4a57b3f487491bbeafaf8083688b16841a4240e5"
 dependencies = [
- "windows-sys 0.59.0",
+ "windows-sys 0.61.2",
 ]
 
 [[package]]
@@ -2423,24 +2442,6 @@ dependencies = [
  "pkg-config",
 ]
 
-[[package]]
-name = "openai_dive"
-version = "1.3.3"
-source = "git+https://github.com/jhqxxx/openai-client.git?tag=1.3.3#f799b7f28180f14e44425bb2df3e5cefc06e31ec"
-dependencies = [
- "bytes",
- "derive_builder",
- "futures",
- "reqwest",
- "reqwest-eventsource",
- "serde",
- "serde_html_form",
- "serde_json",
- "tokio",
- "tokio-stream",
- "tokio-util",
-]
-
 [[package]]
 name = "openssl"
 version = "0.10.74"
@@ -3135,7 +3136,7 @@ dependencies = [
  "errno",
  "libc",
  "linux-raw-sys",
- "windows-sys 0.59.0",
+ "windows-sys 0.61.2",
 ]
 
 [[package]]
@@ -3578,7 +3579,7 @@ dependencies = [
  "getrandom 0.3.4",
  "once_cell",
  "rustix",
- "windows-sys 0.59.0",
+ "windows-sys 0.61.2",
 ]
 
 [[package]]
@@ -4292,7 +4293,7 @@ version = "0.1.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "c2a7b1c03c876122aa43f3020e6c3c3ee5c05081c9a00739faf7503aeba10d22"
 dependencies = [
- "windows-sys 0.59.0",
+ "windows-sys 0.61.2",
 ]
 
 [[package]]

Cargo.toml

@@ -21,7 +21,7 @@ base64 = "0.22.1"
 num = "0.4.3"
 minijinja = "2.12.0"
 tokenizers = "0.22.1"
-openai_dive = { git = "https://github.com/jhqxxx/openai-client.git", version = "1.3.3", tag = "1.3.3", features = ["stream"]}
+aha_openai_dive = {version = "1.3.2", features = ["stream"]}
 uuid = { version = "1.18.1", features = ["v4"]}
 chrono = "0.4.42"
 rocket = "0.5.1"

src/chat_template/mod.rs

@@ -1,6 +1,6 @@
+use aha_openai_dive::v1::resources::chat::ChatCompletionParameters;
 use anyhow::{Result, anyhow};
 use minijinja::{Environment, Value as MiniJinjaValue, context};
-use openai_dive::v1::resources::chat::ChatCompletionParameters;
 
 use crate::utils::string_to_static_str;
 

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 aha_openai_dive::v1::resources::chat::{
-use 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,

src/utils/mod.rs

@@ -3,10 +3,7 @@ pub mod img_utils;
 pub mod tensor_utils;
 pub mod video_utils;
 
-use anyhow::Result;
+use aha_openai_dive::v1::resources::{
-use candle_core::{DType, Device};
-use candle_transformers::generation::{LogitsProcessor, Sampling};
-use openai_dive::v1::resources::{
     chat::{
         ChatCompletionChoice, ChatCompletionChunkChoice, ChatCompletionChunkResponse,
         ChatCompletionResponse, ChatMessage, ChatMessageContent, DeltaChatMessage, DeltaFunction,
@@ -14,6 +11,9 @@ use openai_dive::v1::resources::{
     },
     shared::FinishReason,
 };
+use anyhow::Result;
+use candle_core::{DType, Device};
+use candle_transformers::generation::{LogitsProcessor, Sampling};
 
 pub fn get_device(device: Option<&Device>) -> Device {
     match device {

tests/test_minicpm4.rs

@@ -1,8 +1,8 @@
 use std::{pin::pin, time::Instant};
 
 use aha::models::{GenerateModel, minicpm4::generate::MiniCPMGenerateModel};
+use aha_openai_dive::v1::resources::chat::ChatCompletionParameters;
 use anyhow::Result;
-use openai_dive::v1::resources::chat::ChatCompletionParameters;
 use rocket::futures::StreamExt;
 
 #[test]

tests/test_qwen2_5vl.rs

@@ -1,8 +1,8 @@
 use std::{pin::pin, time::Instant};
 
 use aha::models::{GenerateModel, qwen2_5vl::generate::Qwen2_5VLGenerateModel};
+use aha_openai_dive::v1::resources::chat::ChatCompletionParameters;
 use anyhow::Result;
-use openai_dive::v1::resources::chat::ChatCompletionParameters;
 use rocket::futures::StreamExt;
 
 #[test]

tests/test_qwen3vl.rs

@@ -1,8 +1,8 @@
 use std::{pin::pin, time::Instant};
 
 use aha::models::{GenerateModel, qwen3vl::generate::Qwen3VLGenerateModel};
+use aha_openai_dive::v1::resources::chat::ChatCompletionParameters;
 use anyhow::Result;
-use openai_dive::v1::resources::chat::ChatCompletionParameters;
 use rocket::futures::StreamExt;
 
 #[test]