feat(together-ai): update model YAMLs [bot] (#1362)

Co-authored-by: Truefoundry Models Bot <truefoundry-models-bot@users.noreply.github.com>
Co-authored-by: harshivg <harshiv.ganwani@truefoundry.com>

Author: 3 people

providers/together-ai/Qwen/Qwen3-235B-A22B-Instruct-2507-FP8.yaml

@@ -16,6 +16,8 @@ modalities:
 mode: chat
 model: Qwen/Qwen3-235B-A22B-Instruct-2507-FP8
 provisioning: serverless
+removeParams:
+    - reasoning_effort
 sources:
     - https://www.together.ai/models/qwen3-235b-a22b-instruct-2507-fp8
 status: active

providers/together-ai/pearl-ai/gemma-4-31b-it.yaml

@@ -1,6 +1,6 @@
 costs:
-    - input_cost_per_token: 3.9e-7
-      output_cost_per_token: 9.7e-7
+    - input_cost_per_token: 2.8e-7
+      output_cost_per_token: 8.6e-7
       region: "*"
 features:
     - function_calling

providers/together-ai/vidu/vidu-q3.yaml

@@ -11,8 +11,11 @@ modalities:
 mode: video
 model: vidu/vidu-q3
 provisioning: serverless
+removeParams:
+    - reasoning_effort
 sources:
     - https://www.vidu.com/vidu-q3
+    - https://www.together.ai/pricing
 status: active
 supportedModes:
     - video