@@ -2,6 +2,15 @@
 
 ### Added
 
+- **✨ OTLP: HTTP/JSON export format support**
+  - Added `OTLPJSONExporter` class for exporting spans in JSON format over HTTP
+  - Added `otlp_protocol` configuration option to `OTLPConfig` (supports `"http/protobuf"` or `"http/json"`)
+  - `HoneyHiveOTLPExporter` now supports protocol selection via `protocol` parameter
+  - Protocol can be set via environment variable `HH_OTLP_PROTOCOL` or `OTEL_EXPORTER_OTLP_PROTOCOL`
+  - JSON exporter sends spans with `Content-Type: application/json` header
+  - All attribute values are serialized as strings to allow backend type conversion testing
+  - Files: `src/honeyhive/tracer/processing/otlp_exporter.py`, `src/honeyhive/config/models/otlp.py`, `src/honeyhive/tracer/instrumentation/initialization.py`
+
 - **✨ Experiments: Instrumentors support for evaluate()**
   - `honeyhive.experiments.evaluate()` and `run_experiment()` now accept an `instrumentors` parameter
   - Pass factory functions (e.g., `[lambda: GoogleADKInstrumentor()]`) to automatically instrument third-party libraries
 
@@ -109,6 +109,7 @@ Environment Variables
    # Optional
    export HH_SOURCE="my-application"
    export HH_OTLP_ENABLED="true"  # Enable OTLP export (default: true)
+   export HH_OTLP_PROTOCOL="http/json"  # Use JSON format (default: http/protobuf)
 
 Code Configuration
 ~~~~~~~~~~~~~~~~~~
 
@@ -539,6 +539,99 @@ ExperimentConfig
        experiment_metadata={"model": "gpt-4", "temperature": 0.7}
    )
 
+OTLPConfig
+----------
+
+.. autoclass:: OTLPConfig
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+**OTLP (OpenTelemetry Protocol) export configuration parameters.**
+
+**OTLP Fields:**
+
+.. py:attribute:: otlp_enabled
+   :type: bool
+   :value: True
+
+   Enable OTLP export.
+   
+   **Environment Variable**: ``HH_OTLP_ENABLED``
+
+.. py:attribute:: otlp_endpoint
+   :type: Optional[str]
+   :value: None
+
+   Custom OTLP endpoint URL.
+   
+   **Environment Variable**: ``HH_OTLP_ENDPOINT``
+
+.. py:attribute:: otlp_headers
+   :type: Optional[Dict[str, Any]]
+   :value: None
+
+   OTLP headers in JSON format.
+   
+   **Environment Variable**: ``HH_OTLP_HEADERS`` (JSON string)
+
+.. py:attribute:: otlp_protocol
+   :type: str
+   :value: "http/protobuf"
+
+   OTLP protocol format: ``"http/protobuf"`` (default) or ``"http/json"``.
+   
+   **Environment Variables**: ``HH_OTLP_PROTOCOL`` or ``OTEL_EXPORTER_OTLP_PROTOCOL``
+
+.. py:attribute:: batch_size
+   :type: int
+   :value: 100
+
+   OTLP batch size for performance optimization.
+   
+   **Environment Variable**: ``HH_BATCH_SIZE``
+
+.. py:attribute:: flush_interval
+   :type: float
+   :value: 5.0
+
+   OTLP flush interval in seconds.
+   
+   **Environment Variable**: ``HH_FLUSH_INTERVAL``
+
+.. py:attribute:: max_export_batch_size
+   :type: int
+   :value: 512
+
+   Maximum export batch size.
+   
+   **Environment Variable**: ``HH_MAX_EXPORT_BATCH_SIZE``
+
+.. py:attribute:: export_timeout
+   :type: float
+   :value: 30.0
+
+   Export timeout in seconds.
+   
+   **Environment Variable**: ``HH_EXPORT_TIMEOUT``
+
+**Example Usage:**
+
+.. code-block:: python
+
+   from honeyhive.config.models import OTLPConfig
+   
+   # Use JSON format for OTLP export
+   otlp_config = OTLPConfig(
+       otlp_protocol="http/json",
+       batch_size=200,
+       flush_interval=1.0
+   )
+   
+   # Or via environment variable
+   # export HH_OTLP_PROTOCOL=http/json
+   otlp_config = OTLPConfig()  # Loads from HH_OTLP_PROTOCOL
+
 Environment Variable Integration
 ================================
 
@@ -551,6 +644,7 @@ All configuration models support **automatic environment variable loading** usin
 - **Performance**: ``HH_TIMEOUT``, ``HH_MAX_CONNECTIONS``, ``HH_RATE_LIMIT_*``
 - **Caching**: ``HH_CACHE_ENABLED``, ``HH_CACHE_MAX_SIZE``, ``HH_CACHE_TTL``
 - **Experiments**: ``HH_EXPERIMENT_ID``, ``HH_EXPERIMENT_NAME``
+- **OTLP**: ``HH_OTLP_ENABLED``, ``HH_OTLP_ENDPOINT``, ``HH_OTLP_PROTOCOL``, ``HH_OTLP_HEADERS``, ``HH_BATCH_SIZE``, ``HH_FLUSH_INTERVAL``
 
 **Priority Order:**
 
 
@@ -174,7 +174,7 @@ Processing Module (``tracer/processing/``)
    * - ``context.py``
      - Context injection and extraction
    * - ``otlp_exporter.py``
-     - OTLP exporter configuration
+     - OTLP exporter configuration (supports Protobuf and JSON formats)
    * - ``otlp_profiles.py``
      - OTLP export profiles
    * - ``otlp_session.py``
 
@@ -41,6 +41,19 @@ Environment Profile
    :undoc-members:
    :show-inheritance:
 
+OTLP Exporters
+~~~~~~~~~~~~~~~
+
+.. autoclass:: honeyhive.tracer.processing.otlp_exporter.HoneyHiveOTLPExporter
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+.. autoclass:: honeyhive.tracer.processing.otlp_exporter.OTLPJSONExporter
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
 Infrastructure
 --------------
 
 
@@ -405,6 +405,57 @@ HTTP Configuration
 
    **Range**: 1 - 100
 
+OTLP Configuration
+~~~~~~~~~~~~~~~~~~
+
+.. py:data:: otlp_enabled
+   :type: bool
+   :value: True
+
+   **Description**: Enable OTLP export to HoneyHive backend
+   
+   **Environment Variable**: ``HH_OTLP_ENABLED``
+   
+   **Default**: ``True``
+   
+   **Usage**: Set to ``False`` to disable OTLP export (useful for testing)
+
+.. py:data:: otlp_endpoint
+   :type: Optional[str]
+   :value: None
+
+   **Description**: Custom OTLP endpoint URL
+   
+   **Environment Variable**: ``HH_OTLP_ENDPOINT``
+   
+   **Default**: Auto-configured based on ``server_url``
+   
+   **Example**: ``"https://custom.honeyhive.ai/opentelemetry/v1/traces"``
+
+.. py:data:: otlp_protocol
+   :type: str
+   :value: "http/protobuf"
+
+   **Description**: OTLP protocol format for span export
+   
+   **Environment Variables**: ``HH_OTLP_PROTOCOL`` or ``OTEL_EXPORTER_OTLP_PROTOCOL``
+   
+   **Valid Values**:
+   - ``"http/protobuf"`` (default) - Binary Protobuf format
+   - ``"http/json"`` - JSON format for debugging and backend type conversion testing
+   
+   **Example**: Set ``HH_OTLP_PROTOCOL=http/json`` to use JSON format
+
+.. py:data:: otlp_headers
+   :type: Optional[Dict[str, Any]]
+   :value: None
+
+   **Description**: Additional HTTP headers for OTLP export requests
+   
+   **Environment Variable**: ``HH_OTLP_HEADERS`` (JSON string)
+   
+   **Example**: ``{"X-Custom-Header": "value"}``
+
 Tracing Configuration
 ~~~~~~~~~~~~~~~~~~~~~
 
 
@@ -597,6 +597,10 @@ Span Export Format
 
 **OTLP Format** (OpenTelemetry Protocol):
 
+The SDK supports both Protobuf (default) and JSON formats for OTLP export.
+Set the ``HH_OTLP_PROTOCOL`` environment variable to ``http/json`` to use JSON format,
+or ``http/protobuf`` (default) for Protobuf format.
+
 .. code-block:: json
 
    {
 
@@ -66,6 +66,7 @@ Core Capabilities
 - **Graceful Degradation**: SDK never crashes host application, continues operation on failures
 - **Batch Processing**: Configurable span batching for optimal performance
 - **OTLP Performance Tuning**: Environment variables for batch size and flush interval optimization
+- **OTLP JSON Format**: Support for HTTP/JSON export format via ``HH_OTLP_PROTOCOL=http/json``
 - **Production Optimization**: ``HH_BATCH_SIZE`` and ``HH_FLUSH_INTERVAL`` for fine-tuned performance control
 
 **Development & Quality**:
@@ -373,6 +374,7 @@ All configuration supports the ``HH_*`` prefix pattern:
 - **Authentication**: ``HH_API_KEY``, ``HH_SOURCE``
 - **Operational**: ``HH_TEST_MODE``, ``HH_DEBUG_MODE``, ``HH_DISABLE_TRACING``
 - **Performance**: ``HH_TIMEOUT``, ``HH_MAX_CONNECTIONS``, ``HH_RATE_LIMIT_*``, ``HH_BATCH_SIZE``, ``HH_FLUSH_INTERVAL``
+- **OTLP**: ``HH_OTLP_ENABLED``, ``HH_OTLP_ENDPOINT``, ``HH_OTLP_PROTOCOL``, ``HH_OTLP_HEADERS``
 - **Security**: ``HH_SSL_*``, ``HH_PROXY_*``
 
 **Configuration Hierarchy**:
 
@@ -162,7 +162,8 @@ def list_datapoints(
 
         Args:
             project: Project name to filter by
-            dataset: (Legacy) Dataset ID or name to filter by - use dataset_id or dataset_name instead
+            dataset: (Legacy) Dataset ID or name to filter by - use dataset_id or
+                dataset_name instead
             dataset_id: Dataset ID to filter by (takes precedence over dataset_name)
             dataset_name: Dataset name to filter by
 
@@ -206,7 +207,8 @@ async def list_datapoints_async(
 
         Args:
             project: Project name to filter by
-            dataset: (Legacy) Dataset ID or name to filter by - use dataset_id or dataset_name instead
+            dataset: (Legacy) Dataset ID or name to filter by - use dataset_id or
+                dataset_name instead
             dataset_id: Dataset ID to filter by (takes precedence over dataset_name)
             dataset_name: Dataset name to filter by
 
 
@@ -335,7 +335,8 @@ def list_events(
         Uses the proper /events/export POST endpoint as specified in OpenAPI spec.
 
         Args:
-            event_filters: EventFilter or list of EventFilter objects with filtering criteria
+            event_filters: EventFilter or list of EventFilter objects with
+                filtering criteria
             limit: Maximum number of events to return (default: 100)
             project: Project name to filter by (required by API)
             page: Page number for pagination (default: 1)
@@ -347,8 +348,18 @@ def list_events(
             Filter events by type and status::
 
                 filters = [
-                    EventFilter(field="event_type", operator="is", value="model", type="string"),
-                    EventFilter(field="error", operator="is not", value=None, type="string"),
+                    EventFilter(
+                        field="event_type",
+                        operator="is",
+                        value="model",
+                        type="string",
+                    ),
+                    EventFilter(
+                        field="error",
+                        operator="is not",
+                        value=None,
+                        type="string",
+                    ),
                 ]
                 events = client.events.list_events(
                     event_filters=filters,
@@ -471,7 +482,8 @@ async def list_events_async(
         Uses the proper /events/export POST endpoint as specified in OpenAPI spec.
 
         Args:
-            event_filters: EventFilter or list of EventFilter objects with filtering criteria
+            event_filters: EventFilter or list of EventFilter objects with
+                filtering criteria
             limit: Maximum number of events to return (default: 100)
             project: Project name to filter by (required by API)
             page: Page number for pagination (default: 1)
@@ -483,8 +495,18 @@ async def list_events_async(
             Filter events by type and status::
 
                 filters = [
-                    EventFilter(field="event_type", operator="is", value="model", type="string"),
-                    EventFilter(field="error", operator="is not", value=None, type="string"),
+                    EventFilter(
+                        field="event_type",
+                        operator="is",
+                        value="model",
+                        type="string",
+                    ),
+                    EventFilter(
+                        field="error",
+                        operator="is not",
+                        value=None,
+                        type="string",
+                    ),
                 ]
                 events = await client.events.list_events_async(
                     event_filters=filters,
Original file line number	Diff line number	Diff line change
`@@ -597,6 +597,10 @@ Span Export Format`
`597`	`597`
`598`	`598`	`OTLP Format (OpenTelemetry Protocol):`
`599`	`599`
	`600`	`+The SDK supports both Protobuf (default) and JSON formats for OTLP export.`
	`601`	+Set the ``HH_OTLP_PROTOCOL`` environment variable to ``http/json`` to use JSON format,
	`602`	+or ``http/protobuf`` (default) for Protobuf format.
	`603`	`+`
`600`	`604`	`.. code-block:: json`
`601`	`605`
`602`	`606`	`{`