[Bug]: AWS Bedrock default test curl -> model is outdated

[malinthaprasan]

Please select the area the issue is related to

AI Workspace

Please select the aspect the issue is related to

Aspect/UI (Frontend layouts, components, styling)

Description

➜ wso2apip-ai-gateway-1.2.0 curl -X POST "https://localhost:8444/converseapi/model/amazon.titan-text-express-v1/invoke" -k
-H "Content-Type: application/json"
-H "Authorization: Bearer df19d977bf1ba5a7abab472a9be4ba07f49b500d27e6d7d70eee42d3863dd7d4"
-d '{
"inputText": "Say hello!",
"textGenerationConfig": {
"maxTokenCount": 256,
"temperature": 0.7
}
}'
{"message":"This model version has reached the end of its life. Please refer to the AWS documentation for more details."}%

Steps to Reproduce

Create a bedrock provider and invoke using the curl provided by the UI.

Severity Level of the Issue

Severity/Major (Important functionality is broken. Should be prioritized. Doesn't need immediate attention)

Environment Details (with versions)

No response

[malinthaprasan]

Seems like the example curl is also not consistent.

curl -X POST "https://localhost:8444/converseapi/model/openai.gpt-oss-20b-1:0/invoke" -k  \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer df19d977bf1ba5a7abab472a9be4ba07f49b500d27e6d7d70eee42d3863dd7d4" \
  -d '{
  "inputText": "Say hello!",
  "textGenerationConfig": {
    "maxTokenCount": 256,
    "temperature": 0.7
  }
}'
{"message":"{\"error\":{\"code\":\"validation_error\",\"message\":\"Failed to deserialize the JSON body into the target type: missing field `messages` at line 1 column 137\",\"param\":null,\"type\":\"invalid_request_error\"}}"}%   

The path should be /model/{model}/converse (This is the one available in the OpenAPI as well) - and also the body also needs to change as below.

curl -X POST "https://localhost:8444/converseapi/model/openai.gpt-oss-20b-1:0/converse" -k  \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer df19d977bf1ba5a7abab472a9be4ba07f49b500d27e6d7d70eee42d3863dd7d4" \
  -d '{
    "messages": [
        {
            "role": "user",
            "content": [
                {
                    "text": "Hello, who are you?"
                }
            ]
        }
    ],
    "inferenceConfig": {
        "maxTokens": 512,
        "temperature": 0.7
    }
}'
{"metrics":{"latencyMs":498},"output":{"message":{"content":[{"reasoningContent":{"reasoningText":{"text":"We need to respond as ChatGPT. The user: \"Hello, who are you?\" We respond politely, identifying ourselves. We might ask them what they'd like help with. The conversation is likely about generating some text. Just answer."}}},{"text":"Hello! I’m ChatGPT, a virtual assistant powered by OpenAI’s language model. I’m here to help answer questions, brainstorm ideas, write and edit text, explain concepts, and more. How can I assist you today?"}],"role":"assistant"}},"stopReason":"end_turn","usage":{"inputTokens":73,"outputTokens":104,"serverToolUsage":{},"totalTokens":177}}%