Release/v2.1.2

README.md

@@ -20,7 +20,7 @@ pip install agora-agents
 ## Quick Start
 
 Start with the `Agent` builder: create a client with app credentials, choose your ASR, LLM, and TTS providers, then start a session. Omit vendor API keys for supported Agora-managed models, or provide keys when you want BYOK.
-Set Agora interaction language with `turn_detection.language`; provider-specific STT language values remain under `asr.params`.
+Set Agora interaction language with `turn_detection.language`; provider-specific STT language values remain under `asr.params`. Ares uses only the REST `asr.language` value sourced from `turn_detection.language`.
 
 ```python
 import os

changelog.md

@@ -21,7 +21,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/).
 ### Fixed
 
 - **Managed-provider validation** — AgentKit validation now distinguishes preset-backed providers from BYOK providers so required provider fields are only required when credentials are caller-supplied.
-- **Language placement** — Provider-specific STT language values remain under `asr.params`, while Agora interaction language is emitted separately as `turn_detection.language`.
+- **Language placement** — Provider-specific STT language values remain under `asr.params`; the REST `asr.language` field is populated from `turn_detection.language`.
 
 ## [v2.0.0] — 2026-05-21
 
@@ -114,7 +114,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/).
 
 ### Fixed
 
-- **`AresSTT`** — Removed redundant `language` key from the `params` dict. Language is now emitted only at the top level. `params` is only included when `additional_params` is provided.
+- **`AresSTT`** — Removed redundant `language` key from the `params` dict. Ares only selects the provider; AgentKit populates REST `asr.language` from `turn_detection.language`. `params` is only included when `additional_params` is provided.
 - **`OpenAIRealtime` / `VertexAI` (MLLM)** — Agent-level `greeting` and `failure_message` defaults are now correctly applied when missing in MLLM mode. Previously these values were silently dropped.
 - **`VertexAI` (MLLM)** — `messages` is emitted at the MLLM top level, matching the generated core SDK contract.
 

docs/concepts/vendors.md

@@ -75,7 +75,7 @@ tts = ElevenLabsTTS(
 
 Used with `agent.with_stt()`.
 
-Use `turn_detection.language` for Agora interaction language; it defaults to `en-US`. STT vendor `language` options are serialized under `asr.params` using each provider's own format.
+Use `turn_detection.language` for Agora interaction language; it defaults to `en-US`. STT vendor `language` options are serialized under `asr.params` using each provider's own format. Ares does not take a provider language option; AgentKit uses `turn_detection.language` for REST `asr.language`.
 
 | Class | Provider | Required Parameters |
 |---|---|---|

docs/reference/vendors.md

@@ -318,7 +318,7 @@ The SDK also includes named helpers for the remaining Agora-supported LLM provid
 
 ## STT Vendors
 
-Use `turn_detection.language` for Agora interaction language; it defaults to `en-US`. Provider-specific language values remain under `asr.params` and may use a different format.
+Use `turn_detection.language` for Agora interaction language; it defaults to `en-US`. Provider-specific language values remain under `asr.params` and may use a different format. AgentKit populates REST `asr.language` from `turn_detection.language`.
 
 ### `SpeechmaticsSTT`
 
@@ -336,6 +336,7 @@ Use `turn_detection.language` for Agora interaction language; it defaults to `en
 | `api_key` | `str` | BYOK only | `None` | Deepgram API key. Optional only for Agora-managed `nova-2` and `nova-3`. |
 | `model` | `str` | No | `None` | Model (e.g., `nova-2`) |
 | `language` | `str` | No | `None` | Language code (e.g., `en-US`) |
+| `interaction_language` | `str` | No | `None` | Agora `asr.language` override |
 | `smart_format` | `bool` | No | `None` | Enable smart formatting |
 | `punctuation` | `bool` | No | `None` | Enable punctuation |
 | `additional_params` | `Dict[str, Any]` | No | `None` | Additional parameters |
@@ -396,7 +397,6 @@ For `nova-2` and `nova-3`, omit `api_key` to use Agora-managed credentials. For
 
 | Parameter | Type | Required | Default | Description |
 |---|---|---|---|---|
-| `language` | `str` | No | `None` | Language code |
 | `additional_params` | `Dict[str, Any]` | No | `None` | Additional parameters |
 
 ### `SarvamSTT`

src/agora_agent/agentkit/agent.py

@@ -919,9 +919,10 @@ def to_properties(
         allow_missing_llm = "llm" in allow_missing_categories
         allow_missing_tts = "tts" in allow_missing_categories
 
+        turn_detection_config = self._resolve_turn_detection_config()
         if not skip_asr_validation and (self._stt is not None or not allow_missing_asr):
-            base_kwargs["asr"] = self._resolve_asr_config()
-        base_kwargs["turn_detection"] = self._resolve_turn_detection_config()
+            base_kwargs["asr"] = self._resolve_asr_config(turn_detection_config)
+        base_kwargs["turn_detection"] = turn_detection_config
 
         if skip_vendor_validation:
             return StartAgentsRequestProperties(**base_kwargs)
@@ -941,35 +942,30 @@ def to_properties(
 
     def _resolve_llm_config(self) -> typing.Dict[str, typing.Any]:
         llm_config = dict(self._llm or {})
-        # Agent-level fields take priority over the vendor's defaults.
-        # This matches the TS SDK where agent-level values override vendor config.
-        if self._instructions is not None:
+        if self._instructions is not None and "system_messages" not in llm_config:
             llm_config["system_messages"] = [{"role": "system", "content": self._instructions}]
-        if self._greeting is not None:
+        if self._greeting is not None and "greeting_message" not in llm_config:
             llm_config["greeting_message"] = self._greeting
-        if self._greeting_configs is not None:
+        if self._greeting_configs is not None and "greeting_configs" not in llm_config:
             llm_config["greeting_configs"] = _dump_optional_model(self._greeting_configs)
-        if self._failure_message is not None:
+        if self._failure_message is not None and "failure_message" not in llm_config:
             llm_config["failure_message"] = self._failure_message
-        if self._max_history is not None:
+        if self._max_history is not None and "max_history" not in llm_config:
             llm_config["max_history"] = self._max_history
         return llm_config
 
-    def _resolve_asr_config(self) -> typing.Dict[str, typing.Any]:
+    def _resolve_asr_config(self, turn_detection_config: TurnDetectionConfig) -> typing.Dict[str, typing.Any]:
         asr_config = dict(self._stt or {})
-        asr_config.pop("language", None)
         if not asr_config:
             asr_config["vendor"] = "ares"
+        asr_config["language"] = self._field_value(turn_detection_config, "language")
         return asr_config
 
     def _resolve_turn_detection_config(self) -> TurnDetectionConfig:
-        existing_stt_language = self._stt.get("language") if self._stt is not None else None
         existing_turn_detection_language = self._field_value(self._turn_detection, "language")
         language = (
             existing_turn_detection_language
             if existing_turn_detection_language is not None
-            else existing_stt_language
-            if _is_turn_detection_language(existing_stt_language)
             else DEFAULT_TURN_DETECTION_LANGUAGE
         )
         language = _validate_turn_detection_language(language)

src/agora_agent/agentkit/agent_session.py

@@ -24,6 +24,7 @@
     is_generic_avatar,
     is_heygen_avatar,
     is_live_avatar_avatar,
+    is_rtc_avatar,
     validate_avatar_config,
     validate_tts_sample_rate,
 )
@@ -333,15 +334,15 @@ def _build_start_properties(
             properties["tts"] = self._dump_model(self._agent.tts)
         if self._agent.llm is not None:
             llm = dict(self._agent.llm)
-            if self._agent.instructions is not None:
+            if self._agent.instructions is not None and "system_messages" not in llm:
                 llm["system_messages"] = [{"role": "system", "content": self._agent.instructions}]
-            if self._agent.greeting is not None:
+            if self._agent.greeting is not None and "greeting_message" not in llm:
                 llm["greeting_message"] = self._agent.greeting
-            if self._agent.greeting_configs is not None:
+            if self._agent.greeting_configs is not None and "greeting_configs" not in llm:
                 llm["greeting_configs"] = self._dump_model(self._agent.greeting_configs)
-            if self._agent.failure_message is not None:
+            if self._agent.failure_message is not None and "failure_message" not in llm:
                 llm["failure_message"] = self._agent.failure_message
-            if self._agent.max_history is not None:
+            if self._agent.max_history is not None and "max_history" not in llm:
                 llm["max_history"] = self._agent.max_history
             properties["llm"] = llm
         if self._agent.stt is not None:

src/agora_agent/agentkit/presets.py

@@ -137,7 +137,9 @@ def infer_tts_preset(tts: typing.Optional[typing.Dict[str, typing.Any]]) -> typi
     if vendor == "minimax":
         if params.get("key"):
             return None
-        return _MINIMAX_MODEL_TO_PRESET.get(_normalize_model_name(params.get("model")) or "")
+        # Model is no longer in params for the preset path; fall back to the top-level hint.
+        model = _normalize_model_name(params.get("model")) or _normalize_model_name(tts.get("_minimax_preset_model")) or ""
+        return _MINIMAX_MODEL_TO_PRESET.get(model)
     return None
 
 
@@ -184,6 +186,7 @@ def strip_inferred_preset_fields(properties: typing.Dict[str, typing.Any], infer
             params["group_id"] = None
             params["url"] = None
         tts = {k: v for k, v in {**tts, "params": _omit_none(params)}.items() if v is not None}
+        tts.pop("_minimax_preset_model", None)
 
     return {**properties, "asr": asr, "llm": llm, "tts": tts}
 

src/agora_agent/agentkit/vendors/avatar.py

@@ -206,3 +206,4 @@ def to_config(self) -> Dict[str, Any]:
 
         enable = self.options.enable if self.options.enable is not None else True
         return {"enable": enable, "vendor": "anam", "params": params}
+

src/agora_agent/agentkit/vendors/llm.py

@@ -1,4 +1,4 @@
-from typing import Any, Dict, List, Optional
+from typing import Any, Dict, List, Optional, Union
 
 from pydantic import BaseModel, ConfigDict, Field, model_validator
 
@@ -376,8 +376,14 @@ def to_config(self) -> Dict[str, Any]:
         options = _dump_optional_model(self.options)
         options.pop("project_id", None)
         options.pop("location", None)
+        if not options.get("url"):
+            options["url"] = (
+                f"https://{self.options.location}-aiplatform.googleapis.com/v1/projects/"
+                f"{self.options.project_id}/locations/{self.options.location}/"
+                f"publishers/google/models/{self.options.model}:streamGenerateContent?alt=sse"
+            )
         config = Gemini(**options).to_config()
-        params = dict(config["params"])
+        params = dict(config.get("params") or {})
         params["project_id"] = self.options.project_id
         params["location"] = self.options.location
         config["params"] = params

src/agora_agent/agentkit/vendors/mllm.py

@@ -1,3 +1,4 @@
+import warnings
 from typing import Any, Dict, List, Optional
 
 from pydantic import BaseModel, ConfigDict, Field