feat(providers): add OrcaRouter model catalog, LLM, and ASR support

README.md

@@ -197,9 +197,9 @@ Every item below is one more layer sedimented into a default — a capability yo
 - 🎨 **Style Pack Marketplace** — browse, install, and like community **style packs** from the in-app Marketplace, and publish your own (custom system prompt per pack, switchable by hotkey). Backed by a moderated marketplace backend; uploads are reviewed before they go public.
 - ⚡ **Streaming insertion** — polished text is written to the cursor character by character to reduce perceived latency, with an automatic one-shot-paste fallback. Toggle in Settings → Recording.
 - **Toggle and push-to-talk** recording modes, plus a **MediaPlayPause trigger** so wired-earbud inline controls can start and stop recording. `Esc` cancels at any phase, including polish and insertion.
-- **Cloud ASR**: Volcengine streaming ASR (bigasr), iFlytek realtime ASR (RTASR), Alibaba Cloud Bailian (classic realtime / Qwen3 realtime / Fun-ASR-Flash file transcription), StepFun StepAudio (batch + realtime), Zhipu GLM-ASR, Xiaomi MiMo ASR, ElevenLabs Scribe, OpenAI-compatible batch transcription (OpenAI Whisper / Groq / SiliconFlow SenseVoice / OpenRouter / ZenMux), and Apple Speech (macOS).
+- **Cloud ASR**: Volcengine streaming ASR (bigasr), iFlytek realtime ASR (RTASR), Alibaba Cloud Bailian (classic realtime / Qwen3 realtime / Fun-ASR-Flash file transcription), StepFun StepAudio (batch + realtime), Zhipu GLM-ASR, Xiaomi MiMo ASR, OrcaRouter audio-input Gemini, ElevenLabs Scribe, OpenAI-compatible batch transcription (OpenAI Whisper / Groq / SiliconFlow SenseVoice / OpenRouter / ZenMux), and Apple Speech (macOS).
 - **Local ASR**: bundled Qwen3-ASR (0.6B / 1.7B) via vendored `Open-Less/qwen-asr` (macOS); Windows Foundry Local Whisper and sherpa-onnx (experimental) variants.
-- **Polish providers**: Ark (Volcengine), DeepSeek, OpenAI, Google Gemini, Codex OAuth, SiliconFlow, Atlas Cloud, Xiaomi MiMo, CometAPI, OpenRouter, Alibaba Cloud Coding Plan, CodingPlanX, MiniMax, and StepFun — plus any OpenAI-compatible endpoint you bring.
+- **Polish providers**: Ark (Volcengine), DeepSeek, OpenAI, Google Gemini, Codex OAuth, SiliconFlow, Atlas Cloud, Xiaomi MiMo, CometAPI, OpenRouter, OrcaRouter, Alibaba Cloud Coding Plan, CodingPlanX, MiniMax, and StepFun — plus any OpenAI-compatible endpoint you bring.
 - **Four output modes**: raw, light polish, structured (**AI-prompt mode**), and formal. Plus a **translation hotkey** that converts speech directly into the configured target language ([#43](../../issues/43)).
 - **Selection-ask QA panel** — a separate hotkey opens a floating panel that runs voice Q&A against the highlighted text in any app ([#118](../../issues/118)).
 - **Main window**: Overview / History / Vocab / Style / Marketplace / Settings. Persistent tray icon, plus a mini status capsule that floats on screen and follows the display you are typing on (multi-monitor).

README.zh.md

@@ -202,9 +202,9 @@ OpenLess 只做一件事:**把语音变成可用的书面文字(尤其是 AI 提
 - 🎨 **风格包市场**——在应用内的 Marketplace 浏览、安装、点赞社区**风格包**,并发布自己的(每个包一套自定义系统提示词,可用快捷键切换)。由经过审核的市场后端支撑;上传内容公开前会经过审核。
 - ⚡ **流式插入**——润色后的文本逐字符写入光标以降低感知延迟,并带有自动的一次性粘贴回退。可在 设置 → 录音 中切换。
 - **切换式与按住说话(push-to-talk)** 两种录音模式,外加 **MediaPlayPause 触发**,让有线耳机的线控也能开始 / 停止录音。`Esc` 可在任意阶段取消,包括润色与插入。
-- **云端 ASR**:Volcengine 流式 ASR(bigasr)、讯飞实时语音转写(RTASR)、阿里云百炼(经典实时 / Qwen3 实时 / Fun-ASR-Flash 录音文件)、阶跃星辰 StepAudio(批式 + 实时)、智谱 GLM-ASR、小米 MiMo ASR、ElevenLabs Scribe、OpenAI 兼容批量转写(OpenAI Whisper / Groq / 硅基流动 SenseVoice / OpenRouter / ZenMux),以及 Apple Speech(macOS)。
+- **云端 ASR**:Volcengine 流式 ASR(bigasr)、讯飞实时语音转写(RTASR)、阿里云百炼(经典实时 / Qwen3 实时 / Fun-ASR-Flash 录音文件)、阶跃星辰 StepAudio(批式 + 实时)、智谱 GLM-ASR、小米 MiMo ASR、OrcaRouter 音频输入 Gemini、ElevenLabs Scribe、OpenAI 兼容批量转写(OpenAI Whisper / Groq / 硅基流动 SenseVoice / OpenRouter / ZenMux),以及 Apple Speech(macOS)。
 - **本地 ASR**:通过 vendored 的 `Open-Less/qwen-asr` 内置 Qwen3-ASR(0.6B / 1.7B)(macOS);Windows 上的 Foundry Local Whisper 与 sherpa-onnx(实验性)变体。
-- **润色提供方**:Ark(火山方舟)、DeepSeek、OpenAI、Google Gemini、Codex OAuth、硅基流动、Atlas Cloud、小米 MiMo、CometAPI、OpenRouter、阿里云 Coding Plan、CodingPlanX、MiniMax、StepFun,以及你自带的任意 OpenAI 兼容端点。
+- **润色提供方**:Ark(火山方舟)、DeepSeek、OpenAI、Google Gemini、Codex OAuth、硅基流动、Atlas Cloud、小米 MiMo、CometAPI、OpenRouter、OrcaRouter、阿里云 Coding Plan、CodingPlanX、MiniMax、StepFun,以及你自带的任意 OpenAI 兼容端点。
 - **四种输出模式**:原文、轻度润色、结构化(**AI 提示词模式**)、正式。另有一个**翻译快捷键**,将语音直接转换为所配置的目标语言([#43](../../issues/43))。
 - **选区问答面板**——一个独立快捷键打开浮动面板,针对任意应用中被高亮选中的文本进行语音问答([#118](../../issues/118))。
 - **主窗口**:概览 / 历史 / 词典 / 风格 / 市场 / 设置。常驻托盘图标,以及一个浮于屏幕、并跟随你正在输入的显示器的迷你状态胶囊(多显示器)。

openless-all/app/src-tauri/src/asr/mimo.rs

@@ -20,11 +20,19 @@ const MIMO_MAX_CHUNK_DURATION_MS: u64 = 180_000;
 pub const PROVIDER_ID: &str = "xiaomi-mimo-asr";
 pub const DEFAULT_ENDPOINT: &str = "https://api.xiaomimimo.com/v1";
 pub const DEFAULT_MODEL: &str = "mimo-v2.5-asr";
+pub const ORCAROUTER_PROVIDER_ID: &str = "orcarouter";
+pub const ORCAROUTER_DEFAULT_ENDPOINT: &str = "https://api.orcarouter.ai/v1";
+pub const ORCAROUTER_DEFAULT_MODEL: &str = "google/gemini-2.5-flash";
+const ORCAROUTER_TRANSCRIPTION_PROMPT: &str =
+    "Transcribe this audio verbatim. Return only the transcript, without commentary.";
 
 pub struct MimoBatchASR {
     api_key: String,
     base_url: String,
     model: String,
+    provider_name: &'static str,
+    prompt: Option<&'static str>,
+    audio_data_url: bool,
     buffer: Mutex<Vec<u8>>,
 }
 
@@ -34,6 +42,21 @@ impl MimoBatchASR {
             api_key,
             base_url,
             model,
+            provider_name: "MiMo",
+            prompt: None,
+            audio_data_url: true,
+            buffer: Mutex::new(Vec::new()),
+        }
+    }
+
+    pub fn new_orcarouter(api_key: String, base_url: String, model: String) -> Self {
+        Self {
+            api_key,
+            base_url,
+            model,
+            provider_name: "OrcaRouter",
+            prompt: Some(ORCAROUTER_TRANSCRIPTION_PROMPT),
+            audio_data_url: false,
             buffer: Mutex::new(Vec::new()),
         }
     }
@@ -56,7 +79,7 @@ impl MimoBatchASR {
 
     async fn transcribe_inner(&self, pcm: &[u8]) -> Result<RawTranscript> {
         if self.api_key.trim().is_empty() {
-            anyhow::bail!("MiMo API key missing");
+            anyhow::bail!("{} API key missing", self.provider_name);
         }
 
         let duration_ms = pcm_duration_ms(pcm);
@@ -78,7 +101,7 @@ impl MimoBatchASR {
             .map(|chunk| i16::from_le_bytes([chunk[0], chunk[1]]))
             .collect();
         let wav = encode_wav_16k_mono(&samples);
-        let body = mimo_chat_body(&self.model, &wav);
+        let body = audio_input_chat_body(&self.model, &wav, self.prompt, self.audio_data_url);
         let url = mimo_chat_completions_url(&self.base_url)?;
         let client = crate::net::http();
         let resp = client
@@ -87,15 +110,18 @@ impl MimoBatchASR {
             .json(&body)
             .send()
             .await
-            .context("MiMo ASR HTTP request failed")?;
+            .with_context(|| format!("{} ASR HTTP request failed", self.provider_name))?;
 
         if !resp.status().is_success() {
             let status = resp.status();
             let body = resp.text().await.unwrap_or_default();
-            anyhow::bail!("MiMo ASR API error {}: {}", status, body);
+            anyhow::bail!("{} ASR API error {}: {}", self.provider_name, status, body);
         }
 
-        let json: Value = resp.json().await.context("parse MiMo ASR response")?;
+        let json: Value = resp
+            .json()
+            .await
+            .with_context(|| format!("parse {} ASR response", self.provider_name))?;
         Ok(extract_mimo_text(&json).trim().to_string())
     }
 
@@ -126,22 +152,42 @@ pub fn mimo_chat_completions_url(base_url: &str) -> Result<String> {
 }
 
 pub fn mimo_chat_body(model: &str, wav: &[u8]) -> Value {
-    let audio_data = format!(
-        "data:audio/wav;base64,{}",
-        base64::engine::general_purpose::STANDARD.encode(wav)
-    );
+    audio_input_chat_body(model, wav, None, true)
+}
+
+pub fn orcarouter_chat_body(model: &str, wav: &[u8]) -> Value {
+    audio_input_chat_body(model, wav, Some(ORCAROUTER_TRANSCRIPTION_PROMPT), false)
+}
+
+fn audio_input_chat_body(
+    model: &str,
+    wav: &[u8],
+    prompt: Option<&str>,
+    audio_data_url: bool,
+) -> Value {
+    let encoded = base64::engine::general_purpose::STANDARD.encode(wav);
+    let audio_data = if audio_data_url {
+        format!("data:audio/wav;base64,{encoded}")
+    } else {
+        encoded
+    };
+    let mut content = Vec::with_capacity(if prompt.is_some() { 2 } else { 1 });
+    if let Some(text) = prompt {
+        content.push(serde_json::json!({ "type": "text", "text": text }));
+    }
+    content.push(serde_json::json!({
+        "type": "input_audio",
+        "input_audio": {
+            "data": audio_data,
+            "format": "wav",
+        },
+    }));
     serde_json::json!({
         "model": model,
         "stream": false,
         "messages": [{
             "role": "user",
-            "content": [{
-                "type": "input_audio",
-                "input_audio": {
-                    "data": audio_data,
-                    "format": "wav",
-                },
-            }],
+            "content": content,
         }],
     })
 }
@@ -296,6 +342,21 @@ mod tests {
             .starts_with("data:audio/wav;base64,"));
     }
 
+    #[test]
+    fn orcarouter_body_uses_prompt_and_raw_base64_audio() {
+        let body = orcarouter_chat_body(ORCAROUTER_DEFAULT_MODEL, b"wav");
+        assert_eq!(body["model"], ORCAROUTER_DEFAULT_MODEL);
+        assert_eq!(body["messages"][0]["content"][0]["type"], "text");
+        assert_eq!(
+            body["messages"][0]["content"][0]["text"],
+            ORCAROUTER_TRANSCRIPTION_PROMPT
+        );
+        let audio = &body["messages"][0]["content"][1];
+        assert_eq!(audio["type"], "input_audio");
+        assert_eq!(audio["input_audio"]["format"], "wav");
+        assert_eq!(audio["input_audio"]["data"], "d2F2");
+    }
+
     #[test]
     fn mimo_response_extracts_message_content() {
         let json = serde_json::json!({

openless-all/app/src-tauri/src/commands/credentials.rs

@@ -148,6 +148,7 @@ fn llm_provider_default_endpoint(provider: &str) -> Option<&'static str> {
         "mimo" => Some("https://api.xiaomimimo.com/v1"),
         "cometapi" => Some("https://api.cometapi.com/v1"),
         "openrouterFree" => Some("https://openrouter.ai/api/v1"),
+        "orcarouter" => Some("https://api.orcarouter.ai/v1"),
         "alibabaCoding" => Some("https://coding-intl.dashscope.aliyuncs.com/v1"),
         "codingPlanX" => Some("https://api.codingplanx.ai/v1"),
         "stepfun" => Some("https://api.stepfun.com/v1"),

openless-all/app/src-tauri/src/commands/mod.rs

@@ -251,7 +251,9 @@ mod tests {
     };
     use crate::commands::providers::{
         active_asr_is_keyless_for_validation, asr_transcriptions_url, fetch_provider_models,
-        is_gemini_base_url, models_url, parse_gemini_model_ids, parse_model_ids, ProviderConfig,
+        is_gemini_base_url, is_orcarouter_base_url, models_url, parse_gemini_model_ids,
+        parse_model_ids, parse_orcarouter_asr_model_ids, parse_orcarouter_model_ids,
+        ProviderConfig,
     };
     use crate::commands::settings::{
         parse_latest_beta_from_atom, persist_settings, SettingsWriter,
@@ -686,6 +688,22 @@ mod tests {
         assert!(llm_configured_for_provider("atlascloud", &ready));
     }
 
+    #[test]
+    fn credentials_status_requires_api_key_for_orcarouter() {
+        let keyless = CredentialsSnapshot {
+            ark_endpoint: Some("https://api.orcarouter.ai/v1".into()),
+            ark_model_id: Some("orcarouter/fusion-flash".into()),
+            ..snapshot()
+        };
+        assert!(!llm_configured_for_provider("orcarouter", &keyless));
+
+        let ready = CredentialsSnapshot {
+            ark_api_key: Some("key".into()),
+            ..keyless
+        };
+        assert!(llm_configured_for_provider("orcarouter", &ready));
+    }
+
     impl SettingsWriter for FakeSettingsWriter {
         fn read_settings(&self) -> UserPreferences {
             self.saved.lock().unwrap().clone().unwrap_or_default()
@@ -809,6 +827,51 @@ mod tests {
         assert_eq!(models, vec!["a".to_string(), "b".to_string()]);
     }
 
+    #[test]
+    fn orcarouter_catalog_filters_non_chat_models() {
+        let body = r#"{"data":[
+            {"id":"orcarouter/fusion-flash","supported_endpoint_types":["openai","anthropic"]},
+            {"id":"openai/gpt-5-mini","supported_endpoint_types":["openai"]},
+            {"id":"openai/sora-2","supported_endpoint_types":["openai-video"]},
+            {"id":"openai/text-embedding-3-small","supported_endpoint_types":["embeddings"]},
+            {"id":"legacy/chat-model"}
+        ]}"#;
+        assert_eq!(
+            parse_orcarouter_model_ids(body).unwrap(),
+            vec![
+                "legacy/chat-model".to_string(),
+                "openai/gpt-5-mini".to_string(),
+                "orcarouter/fusion-flash".to_string(),
+            ]
+        );
+        assert!(is_orcarouter_base_url("https://api.orcarouter.ai/v1"));
+        assert!(!is_orcarouter_base_url("https://orcarouter.ai/v1"));
+        assert!(!is_orcarouter_base_url(
+            "https://api.orcarouter.ai.example/v1"
+        ));
+    }
+
+    #[test]
+    fn orcarouter_asr_catalog_keeps_audio_input_gemini_models_only() {
+        let body = r#"{"data":[
+            {"id":"google/gemini-2.5-flash","supported_endpoint_types":["openai"]},
+            {"id":"google/gemini-3.5-flash","supported_endpoint_types":["openai"]},
+            {"id":"google/gemini-2.5-flash-image-preview","supported_endpoint_types":["openai"]},
+            {"id":"google/gemini-2.5-flash-tts","supported_endpoint_types":["openai"]},
+            {"id":"openai/whisper-large-v3","supported_endpoint_types":["openai"]},
+            {"id":"google/gemini-legacy"},
+            {"id":"google/gemini-native","supported_endpoint_types":["google"]}
+        ]}"#;
+        assert_eq!(
+            parse_orcarouter_asr_model_ids(body).unwrap(),
+            vec![
+                "google/gemini-2.5-flash".to_string(),
+                "google/gemini-3.5-flash".to_string(),
+                "google/gemini-legacy".to_string(),
+            ]
+        );
+    }
+
     #[test]
     fn parse_gemini_model_ids_strips_models_prefix_and_dedups() {
         // Google v1beta/models 真实响应的子集——name 字段带 `models/` 前缀，