feat: sync full-stack Docker runtime and UI

backend/.env.example

@@ -4,13 +4,13 @@ LLM_BASE_URL=https://api.openai.com/v1/
 LLM_API_KEY=sk-your-key
 
 # Default model for inline completions
-LLM_MODEL=gpt-4.1-mini
+LLM_MODEL=Nex-N2-mini-mlx-OptiQ-8bit-MTP
 
 # Pro-tier model (defaults to LLM_MODEL if unset)
-PRO_LLM_MODEL=gpt-4.1
+PRO_LLM_MODEL=Nex-N2-mini-mlx-OptiQ-8bit-MTP
 
 # Vision model for OCR
-VLM_MODEL=gpt-4.1-mini
+VLM_MODEL=Nex-N2-mini-mlx-OptiQ-8bit-MTP
 
 # API key for the FastAPI app (change in production)
 API_KEY=your-secret-key-here
@@ -56,10 +56,10 @@ JOB_OCR_CONCURRENCY=1
 JOB_OCR_MAX_QUEUE=8
 JOB_CONVERT_CONCURRENCY=1
 JOB_CONVERT_MAX_QUEUE=8
-JOB_TTS_CONCURRENCY=1
-JOB_TTS_MAX_QUEUE=4
-JOB_ASR_CONCURRENCY=1
-JOB_ASR_MAX_QUEUE=4
+JOB_TTS_CONCURRENCY=4
+JOB_TTS_MAX_QUEUE=16
+JOB_ASR_CONCURRENCY=2
+JOB_ASR_MAX_QUEUE=8
 
 # Timeouts (seconds)
 LLM_COMPLETION_TIMEOUT=600
@@ -88,10 +88,12 @@ RISK_MODEL_CIRCUIT_TTL_SECONDS=300
 RISK_ENFORCE_REDIS_FAIL_CLOSED=false
 
 # Backend-controlled model policy
-RISK_COMPLETION_MODEL=gpt-4.1-mini
-RISK_PRO_MODEL=gpt-4.1
-RISK_VISION_MODEL=gpt-4.1-mini
-RISK_WEB_SEARCH_MODEL=gpt-4.1-mini
+RISK_COMPLETION_MODEL=Nex-N2-mini-mlx-OptiQ-8bit-MTP
+RISK_PRO_MODEL=Nex-N2-mini-mlx-OptiQ-8bit-MTP
+RISK_VISION_MODEL=Nex-N2-mini-mlx-OptiQ-8bit-MTP
+RISK_WEB_SEARCH_MODEL=Nex-N2-mini-mlx-OptiQ-8bit-MTP
+RISK_SPEECH_TTS_MODEL=Qwen3-TTS-12Hz-1.7B-VoiceDesign-8bit
+RISK_SPEECH_ASR_MODEL=Qwen3-ASR-0.6B-8bit
 RISK_COMPLETION_MAX_INPUT_CHARS=24000
 RISK_COMPLETION_MAX_OUTPUT_TOKENS=768
 RISK_COMPLETION_TEMPERATURE=0.4
@@ -104,6 +106,8 @@ RISK_WEB_SEARCH_TEMPERATURE=0.4
 RISK_COMPRESS_MAX_INPUT_CHARS=128000
 RISK_COMPRESS_MAX_OUTPUT_TOKENS=1536
 RISK_OCR_MAX_INPUT_BYTES=104857600
+RISK_SPEECH_TTS_MAX_INPUT_CHARS=4096
+RISK_SPEECH_ASR_MAX_INPUT_BYTES=104857600
 
 # Web search providers
 SEARXNG_BASE_URL=http://searxng:8080
@@ -120,9 +124,18 @@ RISK_PRO_INPUT_COST_PER_1K=0.003
 RISK_PRO_OUTPUT_COST_PER_1K=0.012
 RISK_VISION_INPUT_COST_PER_1K=0.0008
 RISK_VISION_OUTPUT_COST_PER_1K=0.0024
+RISK_SPEECH_TTS_INPUT_COST_PER_1K_CHARS=0
+RISK_SPEECH_TTS_OUTPUT_COST_PER_MINUTE_AUDIO=0
+RISK_SPEECH_ASR_INPUT_COST_PER_MB=0
 
-# Legacy fallback: if LLM_BASE_URL is not set, OLLAMA_HOST will be auto-converted to /v1/ path
-#OLLAMA_HOST=http://localhost:11434
-
-# TTS/ASR settings (see README for full list)
-TTS_ASR_DEVICE=auto
+# Shared speech API settings (uses LLM_BASE_URL + LLM_API_KEY)
+TTS_MODEL_ID=Qwen3-TTS-12Hz-1.7B-VoiceDesign-8bit
+TTS_DEFAULT_INSTRUCTIONS=A clear, natural voice speaking Mandarin Chinese.
+ASR_MODEL_ID=Qwen3-ASR-0.6B-8bit
+TTS_ASR_MAX_TEXT_CHARS=4096
+ASR_MAX_AUDIO_BYTES=104857600
+TTS_ASR_TTS_TIMEOUT_SECONDS=180
+TTS_ASR_ASR_TIMEOUT_SECONDS=300
+TTS_ASR_HEALTHCHECK_TIMEOUT_SECONDS=5
+TTS_ASR_MAX_CONNECTIONS=24
+TTS_ASR_MAX_KEEPALIVE_CONNECTIONS=12