[feat] Add support for processing and uploading multimodal messages

util/opentelemetry-util-genai/CHANGELOG-loongsuite.md

@@ -7,4 +7,5 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## Unreleased
 
-- Add support for memory operations. ([#83](https://github.com/alibaba/loongsuite-python-agent/pull/83))
+- Add multimodal separation and upload support for GenAI utils. ([#94](https://github.com/alibaba/loongsuite-python-agent/pull/94))
+- Add support for memory operations. ([#83](https://github.com/alibaba/loongsuite-python-agent/pull/83))

util/opentelemetry-util-genai/pyproject.toml

@@ -36,6 +36,7 @@ upload = "opentelemetry.util.genai._upload:upload_completion_hook"
 [project.optional-dependencies]
 test = ["pytest>=7.0.0"]
 upload = ["fsspec>=2025.9.0"]
+multimodal_upload = ["httpx", "fsspec>=2025.9.0", "numpy", "soundfile"]
 
 [project.urls]
 Homepage = "https://github.com/open-telemetry/opentelemetry-python-contrib/tree/main/util/opentelemetry-util-genai"

util/opentelemetry-util-genai/src/opentelemetry/util/genai/_extended_semconv/gen_ai_extended_attributes.py

@@ -114,6 +114,16 @@
 Note: This is an Attribute, distinct from OpenTelemetry's Span kind field.
 """
 
+GEN_AI_INPUT_MULTIMODAL_METADATA: Final = "gen_ai.input.multimodal_metadata"
+"""
+The multimodal metadata of the input messages.
+"""
+
+GEN_AI_OUTPUT_MULTIMODAL_METADATA: Final = "gen_ai.output.multimodal_metadata"
+"""
+The multimodal metadata of the output messages.
+"""
+
 
 class GenAiSpanKindValues(Enum):
     AGENT = "AGENT"