ibm-watsonx-gov 1.3.3__cp313-cp313-macosx_11_0_arm64.whl

ibm_watsonx_gov/metrics/utils.py

@@ -0,0 +1,440 @@
+# ----------------------------------------------------------------------------------------------------
+# IBM Confidential
+# Licensed Materials - Property of IBM
+# 5737-H76, 5900-A3Q
+# © Copyright IBM Corp. 2025  All Rights Reserved.
+# US Government Users Restricted Rights - Use, duplication or disclosure restricted by
+# GSA ADPSchedule Contract with IBM Corp.
+# ----------------------------------------------------------------------------------------------------
+
+from typing import Dict, List, Optional
+
+import pandas as pd
+
+
+def mapping_to_df(mapping_data: Dict, execution_count: Optional[int] = 0) -> pd.DataFrame:
+    """
+    Convert mapping data into a pandas DataFrame.
+    If execution_count is given, construct dataframe with values w.r.t to execution count
+    """
+    def generate_rows():
+        row = {}
+        for span, lvl1 in mapping_data.items():
+            if not isinstance(lvl1, dict):
+                row[span] = lvl1
+                continue
+
+            for attr, lvl2 in lvl1.items():
+                if not isinstance(lvl2, dict):
+                    col = f"{span}->{attr}"
+                    row[col] = _get_list_value(lvl2, execution_count)
+                    continue
+
+                for path, value in lvl2.items():
+                    col = f"{span}->{attr}->{path}"
+                    row[col] = _get_list_value(value, execution_count)
+        yield row
+
+    return pd.DataFrame(generate_rows())
+
+
+def _get_list_value(value, execution_count):
+    """Helper function to safely get value from list"""
+    if isinstance(value, list):
+        if execution_count < len(value):
+            return value[execution_count]
+        return value[0] if value else None
+    return value
+
+
+TARGETED_USAGE_TRACE_NAMES = [
+    # openAI
+    "openai.embeddings",
+    "ChatOpenAI.chat",
+    "OpenAI.completion",
+    # IBM
+    "ChatWatsonx.chat",
+    "WatsonxLLM.completion",
+    "WatsonxChatModel.chat",  # wxo span
+    # Azure
+    "AzureChatOpenAI.chat",
+    "AzureOpenAI.completion",
+    # AWS
+    "ChatBedrock.chat",
+    "ChatBedrockConverse.chat",
+    # Google
+    "ChatVertexAI.chat",
+    "VertexAI.completion",
+    # Anthropic
+    "ChatAnthropic.chat",
+    "ChatAnthropicMessages.chat",
+    # TODO: Add attributes for other frameworks as well.
+]
+ONE_M = 1000000
+
+LAST_UPDATED = "2025-01-21"  # Date when pricing was last verified
+
+# ref: https://platform.openai.com/docs/pricing
+OPENAI_COSTS = {  # Costs per 1M tokens
+    # GPT-5 Series
+    "gpt-5": {"input": 2.50, "output": 20.0},
+    "gpt-5.1": {"input": 2.50, "output": 20.0},
+    "gpt-5-mini": {"input": 0.45, "output": 3.60},
+    "gpt-5-nano": {"input": 0.05, "output": 0.40},
+    "gpt-5-pro": {"input": 15.0, "output": 120.0},
+
+    # GPT-4.1 Series
+    "gpt-4.1": {"input": 3.50, "output": 14.0},
+    "gpt-4.1-mini": {"input": 0.70, "output": 2.80},
+    "gpt-4.1-nano": {"input": 0.20, "output": 0.80},
+
+    # GPT-4o Series
+    "gpt-4o": {"input": 2.5, "output": 10.0},
+    "gpt-4o-2024-05-13": {"input": 5.0, "output": 15.0},
+    "gpt-4o-2024-08-06": {"input": 3.75, "output": 15.0},
+    "gpt-4o-mini": {"input": 0.15, "output": 0.60},
+    "gpt-4o-mini-2024-07-18": {"input": 0.30, "output": 1.20},
+    "chatgpt-4o-latest": {"input": 5.0, "output": 15.0},
+
+    # GPT-4 Series (Legacy)
+    "gpt-4": {"input": 30.0, "output": 60.0},
+    "gpt-4-turbo": {"input": 10.0, "output": 30.0},
+    "gpt-4-turbo-2024-04-09": {"input": 10.0, "output": 30.0},
+    "gpt-4-0125-preview": {"input": 10.0, "output": 30.0},
+    "gpt-4-1106-preview": {"input": 10.0, "output": 30.0},
+    "gpt-4-0613": {"input": 30.0, "output": 60.0},
+    "gpt-4-32k": {"input": 60.0, "output": 120.0},
+
+    # O-Series (Reasoning)
+    "o1": {"input": 15.0, "output": 60.0},
+    "o1-pro": {"input": 150.0, "output": 600.0},
+    "o1-mini": {"input": 1.10, "output": 4.40},
+    "o3": {"input": 3.50, "output": 14.0},
+    "o3-pro": {"input": 20.0, "output": 80.0},
+    "o3-mini": {"input": 1.10, "output": 4.40},
+    "o4-mini": {"input": 2.0, "output": 8.0},
+    "o3-deep-research": {"input": 10.0, "output": 40.0},
+    "o4-mini-deep-research": {"input": 2.0, "output": 8.0},
+
+    # GPT-3.5 Series
+    "gpt-3.5-turbo": {"input": 0.50, "output": 1.50},
+    "gpt-3.5-turbo-0125": {"input": 0.50, "output": 1.50},
+    "gpt-3.5-turbo-1106": {"input": 1.0, "output": 2.0},
+    "gpt-3.5-turbo-instruct": {"input": 1.50, "output": 2.0},
+
+    # Base Models
+    "davinci-002": {"input": 2.0, "output": 2.0},
+    "babbage-002": {"input": 0.40, "output": 0.40},
+
+    # Embeddings
+    "text-embedding-3-large": {"input": 0.13, "output": 0.0},
+    "text-embedding-3-small": {"input": 0.02, "output": 0.0},
+    "text-embedding-ada-002": {"input": 0.10, "output": 0.0},
+}
+
+# ref: https://docs.anthropic.com/en/docs/about-claude/models/overview#model-pricing
+ANTHROPIC_COSTS = {  # Costs per 1M tokens
+    # Opus 4 Series
+    "claude-opus-4-1": {"input": 15.0, "output": 75.0},
+    "claude-opus-4-0": {"input": 15.0, "output": 75.0},
+    "claude-opus-4-20250514": {"input": 15.0, "output": 75.0},
+    "anthropic.claude-opus-4-20250514-v1:0": {"input": 15.0, "output": 75.0},
+    "claude-opus-4@20250514": {"input": 15.0, "output": 75.0},
+
+    # Sonnet 4.5 Series (tiered pricing based on prompt size)
+    "claude-sonnet-4-5": {"input": 3.0, "output": 15.0},  # ≤200K tokens
+    "claude-sonnet-4-5-large": {"input": 6.0, "output": 22.50},  # >200K tokens
+
+    # Sonnet 4 Series
+    "claude-sonnet-4-0": {"input": 3.0, "output": 15.0},
+    "claude-sonnet-4-20250514": {"input": 3.0, "output": 15.0},
+    "anthropic.claude-sonnet-4-20250514-v1:0": {"input": 3.0, "output": 15.0},
+    "claude-sonnet-4@20250514": {"input": 3.0, "output": 15.0},
+
+    # Sonnet 3.7 Series
+    "claude-3-7-sonnet-latest": {"input": 3.0, "output": 15.0},
+    "claude-3-7-sonnet-20250219": {"input": 3.0, "output": 15.0},
+    "anthropic.claude-3-7-sonnet-20250219-v1:0": {"input": 3.0, "output": 15.0},
+    "claude-3-7-sonnet@20250219": {"input": 3.0, "output": 15.0},
+
+    # Sonnet 3.5 Series
+    "claude-3-5-sonnet-latest": {"input": 3.0, "output": 15.0},
+    "claude-3-5-sonnet-20241022": {"input": 3.0, "output": 15.0},
+    "anthropic.claude-3-5-sonnet-20241022-v2:0": {"input": 3.0, "output": 15.0},
+    "claude-3-5-sonnet-v2@20241022": {"input": 3.0, "output": 15.0},
+
+    # Haiku 4.5 Series
+    "claude-haiku-4-5": {"input": 1.0, "output": 5.0},
+
+    # Haiku 3.5 Series
+    "claude-3-5-haiku-latest": {"input": 0.80, "output": 4.0},
+    "claude-3-5-haiku-20241022": {"input": 0.80, "output": 4.0},
+    "anthropic.claude-3-5-haiku-20241022-v1:0": {"input": 0.80, "output": 4.0},
+    "claude-3-5-haiku@20241022": {"input": 0.80, "output": 4.0},
+
+    # Haiku 3 Series
+    "claude-3-haiku-20240307": {"input": 0.25, "output": 1.25},
+    "anthropic.claude-3-haiku-20240307-v1:0": {"input": 0.25, "output": 1.25},
+    "claude-3-haiku@20240307": {"input": 0.25, "output": 1.25},
+}
+
+# ref: https://cloud.google.com/vertex-ai/generative-ai/pricing
+GOOGLE_COSTS = {  # Costs per 1M tokens
+    # Gemini 3 Series
+    "gemini-3-pro-preview": {"input": 2.0, "output": 12.0},  # ≤200K tokens
+    # >200K tokens
+    "gemini-3-pro-preview-large": {"input": 4.0, "output": 18.0},
+
+    # Gemini 2.5 Series
+    "gemini-2.5-pro": {"input": 1.25, "output": 10.0},  # ≤200K tokens
+    "gemini-2.5-pro-large": {"input": 2.50, "output": 15.0},  # >200K tokens
+    "gemini-2.5-flash": {"input": 0.30, "output": 2.50},
+    "gemini-2.5-flash-lite": {"input": 0.1, "output": 0.4},
+
+    # Gemini 2.0 Series
+    "gemini-2.0-flash-001": {"input": 0.15, "output": 0.6},
+    "gemini-2.0-flash-lite-001": {"input": 0.075, "output": 0.3},
+}
+
+# ref: https://mistral.ai/pricing#api-pricing
+# ref: https://aws.amazon.com/bedrock/pricing
+# ref: https://cloud.google.com/vertex-ai/generative-ai/pricing
+MISTRAL_COSTS = {  # Costs per 1M tokens
+    # Pixtral Series (Vision)
+    "pixtral-large-latest": {"input": 2.0, "output": 6.0},
+    "mistral.pixtral-large-2502-v1:0": {"input": 2.0, "output": 6.0},
+    "pixtral-12b": {"input": 0.15, "output": 0.15},
+
+    # Mistral Large Series
+    "mistral-large-latest": {"input": 2.0, "output": 6.0},
+    "mistral.mistral-large-2407-v1:0": {"input": 2.0, "output": 6.0},
+    "mistralai/mistral-large-2411@001": {"input": 2.0, "output": 6.0},
+    "mistral.mistral-large-2402-v1:0": {"input": 4.0, "output": 12.0},
+
+    # Mistral Medium Series
+    "mistral-medium-3": {"input": 0.4, "output": 2.0},
+    "mistral-medium-latest": {"input": 0.4, "output": 2.0},
+
+    # Mistral Small Series
+    "mistral-small-3-2": {"input": 0.1, "output": 0.3},
+    "mistral-small-latest": {"input": 0.1, "output": 0.3},
+    "mistralai/mistral-small-2503@001": {"input": 0.1, "output": 0.3},
+    "mistral.mistral-small-2402-v1:0": {"input": 1.0, "output": 3.0},
+
+    # Magistral Series (Reasoning)
+    "magistral-medium-latest": {"input": 2.0, "output": 5.0},
+    "magistral-small-latest": {"input": 0.5, "output": 1.5},
+
+    # Devstral Series (Coding)
+    "devstral-medium-2507": {"input": 0.4, "output": 2.0},
+    "devstral-small-2507": {"input": 0.1, "output": 0.3},
+
+    # Codestral Series
+    "codestral-latest": {"input": 0.3, "output": 0.9},
+
+    # Ministral Series (Edge)
+    "ministral-8b-latest": {"input": 0.1, "output": 0.1},
+    "ministral-3b-latest": {"input": 0.04, "output": 0.04},
+
+    # Open Mistral Series
+    "open-mistral-7b": {"input": 0.25, "output": 0.25},
+    "mistral.mistral-7b-instruct-v0:2": {"input": 0.15, "output": 0.2},
+    "open-mistral-nemo": {"input": 0.15, "output": 0.15},
+
+    # Open Mixtral Series
+    "open-mixtral-8x7b": {"input": 0.7, "output": 0.7},
+    "mistral.mixtral-8x7b-instruct-v0:1": {"input": 0.45, "output": 0.7},
+    "open-mixtral-8x22b": {"input": 2.0, "output": 6.0},
+
+    # Embeddings
+    "mistral-embed": {"input": 0.1, "output": 0.0},
+    "codestral-embed-2505": {"input": 0.15, "output": 0.0},
+}
+
+# ref: https://aws.amazon.com/bedrock/pricing
+COHERE_COSTS = {  # Costs per 1M tokens
+    "command-r": {"input": 0.5, "output": 1.5},
+    "cohere.command-r-v1:0": {"input": 0.5, "output": 1.5},
+    "command-r-plus": {"input": 3.0, "output": 15},
+    "cohere.command-r-plus-v1:0": {"input": 3.0, "output": 15},
+    "command-light": {"input": 0.3, "output": 0.6},
+    "cohere.command-light-text-v14": {"input": 0.3, "output": 0.6},
+    "command": {"input": 1.0, "output": 2.0},
+    "cohere.command-text-v14": {"input": 1.0, "output": 2.0},
+}
+
+# ref: https://www.ai21.com/pricing
+# ref: https://aws.amazon.com/bedrock/pricing
+# ref: https://cloud.google.com/vertex-ai/generative-ai/pricing
+AI21_COSTS = {  # Costs per 1M tokens
+    "jamba-large": {"input": 2.0, "output": 8.0},
+    "ai21.jamba-1-5-large-v1:0": {"input": 2.0, "output": 8.0},
+    "ai21/jamba-1.5-large@001": {"input": 2.0, "output": 8.0},
+    "jamba-mini": {"input": 0.2, "output": 0.4},
+    "ai21.jamba-1-5-mini-v1:0": {"input": 0.2, "output": 0.4},
+    "ai21/jamba-1.5-mini@001": {"input": 0.2, "output": 0.4},
+}
+
+# ref: https://www.ibm.com/products/watsonx-ai/pricing
+IBM_WATSONX_COSTS = {  # Costs per 1M tokens
+    # IBM Granite Series
+    "ibm/granite-4-h-small": {"input": 0.06, "output": 0.25},
+    "ibm/granite-vision-3-2-2b": {"input": 0.10, "output": 0.10},
+    "ibm/granite-3-2b-instruct": {"input": 0.10, "output": 0.10},
+    "ibm/granite-3-8b-instruct": {"input": 0.20, "output": 0.20},
+    "ibm/granite-guardian-3-8b": {"input": 0.20, "output": 0.20},
+    "ibm/granite-8b-code-instruct": {"input": 0.20, "output": 0.20},
+    "ibm/granite-8b-japanese": {"input": 0.60, "output": 0.60},
+    "ibm/granite-3-2-8b-instruct": {"input": 0.20, "output": 0.20},
+    "ibm/granite-timeseries-ttm-r2": {"input": 0.38, "output": 0.38},
+
+    # Meta Llama Series
+    "meta-llama/llama-3-2-1b-instruct": {"input": 0.1, "output": 0.1},
+    "meta-llama/llama-3-2-3b-instruct": {"input": 0.15, "output": 0.15},
+    "meta-llama/llama-3-2-11b-vision-instruct": {"input": 0.35, "output": 0.35},
+    "meta-llama/llama-3-2-90b-vision-instruct": {"input": 2.0, "output": 2.0},
+    "meta-llama/llama-3-3-70b-instruct": {"input": 0.71, "output": 0.71},
+    "meta-llama/llama-3-405b-instruct": {"input": 5.0, "output": 16.0},
+    "meta-llama/llama-4-maverick-17b-128e-instruct-fp": {"input": 0.35, "output": 1.4},
+    "meta-llama/llama-guard-3-11b-vision": {"input": 0.35, "output": 0.35},
+
+    # Mistral Series on Watsonx
+    "mistralai/mistral-small-3-1-24b-instruct-2503": {"input": 0.1, "output": 0.3},
+    "mistralai/mistral-medium-2505": {"input": 3.0, "output": 10.0},
+
+    # Other Models
+    "core42/jais-13b-chat": {"input": 1.8, "output": 1.8},
+    "sdaia/allam-1-13b-instruct": {"input": 1.8, "output": 1.8},
+
+    # IBM Granite Embeddings
+    "ibm/granite-embedding-107m-multilingual": {"input": 0.10, "output": 0.0},
+    "ibm/granite-embedding-278m-multilingual": {"input": 0.10, "output": 0.0},
+
+    # IBM Slate Retrieval Models
+    "ibm/slate-125m-english-rtrvr": {"input": 0.10, "output": 0.0},
+    "ibm/slate-125m-english-rtrvr-v2": {"input": 0.10, "output": 0.0},
+    "ibm/slate-30m-english-rtrvr": {"input": 0.10, "output": 0.0},
+    "ibm/slate-30m-english-rtrvr-v2": {"input": 0.10, "output": 0.0},
+
+    # Third-Party Embeddings
+    "intfloat/multilingual-e5-large": {"input": 0.10, "output": 0.0},
+    "sentence-transformers/all-minilm-l12-v2": {"input": 0.10, "output": 0.0},
+    "sentence-transformers/all-minilm-l6-v2": {"input": 0.10, "output": 0.0},
+
+    # OpenAI
+    "openai/gpt-oss-120b": {"input": 0.15, "output": 0.60}
+}
+
+# ref: https://aws.amazon.com/bedrock/pricing/ (DeepSeek tab)
+DEEPSEEK_COSTS = {  # Costs per 1M tokens
+    # DeepSeek models (Standard Tier - AWS Bedrock)
+    "deepseek-r1": {"input": 1.35, "output": 5.40},
+    "deepseek-v3.1": {"input": 0.58, "output": 1.68},
+}
+
+# ref: https://docs.x.ai/docs/models
+XAI_COSTS = {  # Costs per 1M tokens
+    # Grok 4.1 Series
+    "grok-4-1-fast-reasoning": {"input": 0.20, "output": 0.50},
+    "grok-4-1-fast-non-reasoning": {"input": 0.20, "output": 0.50},
+
+    # Grok 4 Series
+    "grok-4-fast-reasoning": {"input": 0.20, "output": 0.50},
+    "grok-4-fast-non-reasoning": {"input": 0.20, "output": 0.50},
+    "grok-4-0709": {"input": 3.0, "output": 15.0},
+
+    # Grok 3 Series
+    "grok-3": {"input": 3.0, "output": 15.0},
+    "grok-3-mini": {"input": 0.30, "output": 0.50},
+
+    # Grok 2 Series
+    # Both us-east-1 and eu-west-1
+    "grok-2-vision-1212": {"input": 2.0, "output": 10.0},
+
+    # Grok Code
+    "grok-code-fast-1": {"input": 0.20, "output": 1.50},
+
+    # Image Generation (per image)
+    "grok-2-image-1212": {"input": 0.0, "output": 0.07},  # $0.07 per image
+}
+
+# ref: https://docs.perplexity.ai/getting-started/pricing
+PERPLEXITY_COSTS = {  # Costs per 1M tokens
+    # Sonar Models
+    "sonar": {"input": 1.0, "output": 1.0},
+    "sonar-pro": {"input": 3.0, "output": 15.0},
+    "sonar-reasoning": {"input": 1.0, "output": 5.0},
+    "sonar-reasoning-pro": {"input": 2.0, "output": 8.0},
+    # Plus $2/1M citation tokens, $5/1K search queries, $3/1M reasoning tokens
+    "sonar-deep-research": {"input": 2.0, "output": 8.0},
+}
+
+# ref: https://groq.com/pricing
+GROQ_COSTS = {  # Costs per 1M tokens
+    # GPT OSS Series
+    "gpt-oss-20b": {"input": 0.075, "output": 0.30},
+    "gpt-oss-safeguard-20b": {"input": 0.075, "output": 0.30},
+    "gpt-oss-120b": {"input": 0.15, "output": 0.60},
+
+    # Kimi Series
+    "kimi-k2-0905-1t": {"input": 1.0, "output": 3.0},
+
+    # Llama 4 Series
+    "llama-4-scout-17b-16e": {"input": 0.11, "output": 0.34},
+    "llama-4-maverick-17b-128e": {"input": 0.20, "output": 0.60},
+    "llama-guard-4-12b": {"input": 0.20, "output": 0.20},
+
+    # Qwen Series
+    "qwen3-32b": {"input": 0.29, "output": 0.59},
+
+    # Llama 3.3 Series
+    "llama-3.3-70b-versatile": {"input": 0.59, "output": 0.79},
+
+    # Llama 3.1 Series
+    "llama-3.1-8b-instant": {"input": 0.05, "output": 0.08},
+}
+
+# Combined metadata for backward compatibility
+COST_METADATA = {
+    **OPENAI_COSTS,
+    **ANTHROPIC_COSTS,
+    **GOOGLE_COSTS,
+    **MISTRAL_COSTS,
+    **COHERE_COSTS,
+    **AI21_COSTS,
+    **IBM_WATSONX_COSTS,
+    **DEEPSEEK_COSTS,
+    **XAI_COSTS,
+    **PERPLEXITY_COSTS,
+    **GROQ_COSTS,
+}
+
+
+def calculate_cost(usage_data: List[dict]) -> float:
+    """Calculate cost for given list of usage.
+    [
+        {"model": ..., "total_prompt_tokens": ...,  "total_completion_tokens": ...},
+        ...
+    ]
+    """
+    total_cost = 0.0
+
+    for data in usage_data:
+        model = data["model"].lower()
+
+        try:
+            model_pricing = COST_METADATA[model]
+        except KeyError:
+            return 0
+            # raise ValueError(
+            #     f"Pricing not available for {model}")
+
+        # Calculate costs (per 1M tokens)
+        input_cost = (data.get("total_prompt_tokens", 0) /
+                      ONE_M) * model_pricing["input"]
+        output_cost = (data.get("total_completion_tokens", 0) / ONE_M) * model_pricing[
+            "output"
+        ]
+        total_cost += input_cost + output_cost
+
+    return total_cost

ibm_watsonx_gov/metrics/violence/__init__.py

@@ -0,0 +1,8 @@
+# ----------------------------------------------------------------------------------------------------
+# IBM Confidential
+# Licensed Materials - Property of IBM
+# 5737-H76, 5900-A3Q
+# © Copyright IBM Corp. 2025  All Rights Reserved.
+# US Government Users Restricted Rights - Use, duplication or disclosure restricted by
+# GSA ADPSchedule Contract with IBM Corp.
+# ----------------------------------------------------------------------------------------------------

ibm_watsonx_gov/metrics/violence/violence_decorator.py

@@ -0,0 +1,60 @@
+# ----------------------------------------------------------------------------------------------------
+# IBM Confidential
+# Licensed Materials - Property of IBM
+# 5737-H76, 5900-A3Q
+# © Copyright IBM Corp. 2025  All Rights Reserved.
+# US Government Users Restricted Rights - Use, duplication or disclosure restricted by
+# GSA ADPSchedule Contract with IBM Corp.
+# ----------------------------------------------------------------------------------------------------
+
+from functools import partial
+from typing import Callable, Optional
+
+from wrapt import decorator
+
+from ibm_watsonx_gov.config.agentic_ai_configuration import \
+    AgenticAIConfiguration
+from ibm_watsonx_gov.entities.enums import EvaluatorFields
+from ibm_watsonx_gov.entities.metric import GenAIMetric
+from ibm_watsonx_gov.metrics.base_metric_decorator import BaseMetricDecorator
+from ibm_watsonx_gov.metrics.violence.violence_metric import ViolenceMetric
+
+
+class ViolenceDecorator(BaseMetricDecorator):
+
+    def evaluate_violence(self,
+                          func: Optional[Callable] = None,
+                          *,
+                          configuration: Optional[AgenticAIConfiguration] = None,
+                          metrics: list[GenAIMetric] = []
+                          ) -> dict:
+        """
+        An evaluation decorator for computing violence on an agentic node via granite guardian.
+        """
+        if func is None:
+            return partial(self.evaluate_violence, configuration=configuration, metrics=metrics)
+
+        if not metrics:
+            metrics = [ViolenceMetric()]
+
+        @decorator
+        def wrapper(func, instance, args, kwargs):
+
+            try:
+                self.validate(func=func, metrics=metrics,
+                              valid_metric_types=(ViolenceMetric))
+
+                metric_inputs = [EvaluatorFields.INPUT_FIELDS]
+
+                original_result = self.compute_helper(func=func, args=args, kwargs=kwargs,
+                                                      configuration=configuration,
+                                                      metrics=metrics,
+                                                      metric_inputs=metric_inputs,
+                                                      metric_outputs=[])
+
+                return original_result
+            except Exception as ex:
+                raise Exception(
+                    f"There was an error while evaluating violence on {func.__name__},") from ex
+
+        return wrapper(func)

ibm_watsonx_gov/metrics/violence/violence_metric.py

@@ -0,0 +1,103 @@
+# ----------------------------------------------------------------------------------------------------
+# IBM Confidential
+# Licensed Materials - Property of IBM
+# 5737-H76, 5900-A3Q
+# © Copyright IBM Corp. 2025  All Rights Reserved.
+# US Government Users Restricted Rights - Use, duplication or disclosure restricted by
+# GSA ADPSchedule Contract with IBM Corp.
+# ----------------------------------------------------------------------------------------------------
+
+from typing import Annotated, Literal
+
+import pandas as pd
+from pydantic import Field
+
+from ibm_watsonx_gov.config.gen_ai_configuration import GenAIConfiguration
+from ibm_watsonx_gov.entities.enums import MetricGroup, TaskType
+from ibm_watsonx_gov.entities.evaluation_result import AggregateMetricResult
+from ibm_watsonx_gov.entities.metric import GenAIMetric
+from ibm_watsonx_gov.entities.metric_threshold import MetricThreshold
+from ibm_watsonx_gov.providers.detectors_provider import DetectorsProvider
+from ibm_watsonx_gov.utils.async_util import run_in_event_loop
+from ibm_watsonx_gov.utils.validation_util import validate_input
+
+VIOLENCE = "violence"
+
+
+class ViolenceMetric(GenAIMetric):
+    """
+    Defines the Violence metric class.
+
+    The Violence metric measures the risk of content promoting or describing physical harm to individuals or groups. This includes depictions of assault, self-harm, or creation of a hostile environment through threats or intimidation.
+    It is computed using the granite guardian model.
+
+    Examples:
+        1. Create Violence metric with default parameters and compute using metrics evaluator.
+            .. code-block:: python
+
+                metric = ViolenceMetric()
+                result = MetricsEvaluator().evaluate(data={"input_text": "...", metrics=[metric])
+
+        2. Create Violence metric with a custom threshold.
+            .. code-block:: python
+
+                threshold  = MetricThreshold(type="lower_limit", value=0.5)
+                metric = ViolenceMetric(threshold=threshold)
+    """
+    name: Annotated[Literal["violence"],
+                    Field(title="Name",
+                          description="The violence metric name.",
+                          default=VIOLENCE, frozen=True)]
+    display_name: Annotated[Literal["Violence"],
+                            Field(title="Display Name",
+                                  description="The violence metric display name.",
+                                  default="Violence", frozen=True)]
+    method: Annotated[Literal["granite_guardian"],
+                      Field(title="Method",
+                            description="The method used to compute harm metric.",
+                            default="granite_guardian")]
+    tasks: Annotated[list[TaskType],
+                     Field(title="Tasks",
+                           description="The list of supported tasks.",
+                           default=TaskType.values(), frozen=True)]
+    thresholds: Annotated[list[MetricThreshold],
+                          Field(title="Thresholds",
+                                description="The metric thresholds.",
+                                default=[MetricThreshold(type="upper_limit", value=0.5)])]
+    group: Annotated[MetricGroup,
+                     Field(title="Group",
+                           description="The metric group.",
+                           default=MetricGroup.CONTENT_SAFETY, frozen=True)]
+
+    async def evaluate_async(
+        self,
+        data: pd.DataFrame | dict,
+        configuration: GenAIConfiguration,
+        **kwargs
+    ) -> list[AggregateMetricResult]:
+
+        validate_input(data.columns.to_list(), configuration)
+        kwargs["detector_params"] = {"risk_name": VIOLENCE}
+        provider = DetectorsProvider(configuration=configuration,
+                                     metric_name=self.name,
+                                     metric_display_name=self.display_name,
+                                     metric_method=self.method,
+                                     metric_group=self.group,
+                                     thresholds=self.thresholds,
+                                     **kwargs)
+        aggregated_metric_result = await provider.evaluate_async(data=data)
+        return aggregated_metric_result
+
+    def evaluate(
+        self,
+        data: pd.DataFrame | dict,
+        configuration: GenAIConfiguration,
+        **kwargs,
+    ):
+        # If ran in sync mode, block until it is done
+        return run_in_event_loop(
+            self.evaluate_async,
+            data=data,
+            configuration=configuration,
+            **kwargs,
+        )

ibm_watsonx_gov/prompt_evaluator/__init__.py

@@ -0,0 +1,9 @@
+# ----------------------------------------------------------------------------------------------------
+# IBM Confidential
+# OCO Source Materials
+# 5900-A3Q, 5737-H76
+# Copyright IBM Corp. 2025
+# The source code for this program is not published or other-wise divested of its trade
+# secrets, irrespective of what has been deposited with the U.S.Copyright Office.
+# ----------------------------------------------------------------------------------------------------
+from .prompt_evaluator import PromptEvaluator

ibm_watsonx_gov/prompt_evaluator/impl/__init__.py

@@ -0,0 +1,8 @@
+# ----------------------------------------------------------------------------------------------------
+# IBM Confidential
+# OCO Source Materials
+# 5900-A3Q, 5737-H76
+# Copyright IBM Corp. 2025
+# The source code for this program is not published or other-wise divested of its trade
+# secrets, irrespective of what has been deposited with the U.S.Copyright Office.
+# ----------------------------------------------------------------------------------------------------