tokenbee-sdk 1.2.2__tar.gz → 1.3.0__tar.gz

{tokenbee_sdk-1.2.2 → tokenbee_sdk-1.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tokenbee-sdk
-Version: 1.2.2
+Version: 1.3.0
 Summary: Official Python SDK for TokenBee LLM inference gateway and observability.
 Author: TokenBee Inc.
 Author-email: "TokenBee Inc." <founders@tokenbee.io>
@@ -73,20 +73,26 @@ TokenBee is a **stateless** gateway. We do not store your LLM provider API keys
 
 ### Compression Control
 
-You can specify the compression rate and method per request:
+You can specify the compression rate and method per request. TokenBee uses an intelligent semantic engine to reduce token usage while preserving meaning.
 
 ```python
 response = client.send(
     model=TokenBeeModel.ANTHROPIC_CLAUDE_3_5_SONNET,
     input={
         "messages": [...],
-        "compression": "on",
-        "rate": CompressionRate.HIGH,
+        "compression": "auto",      # "auto" (default), "on", or "off"
+        "rate": CompressionRate.HIGH, # MEDIUM (0.5), HIGH (0.33), etc.
         "privacy": True
     }
 )
 ```
 
+- **`compression`**: Set to `"auto"` to let TokenBee decide when to compress, or `"off"` to bypass the compression engine entirely for high-precision tasks.
+- **`rate`**: Controls the aggressiveness of compression. `HIGH` aims for ~67% token reduction.
+- **`sessionId`**: (Optional) String ID to group multiple requests into a single replayable session in the dashboard.
+- **`userId`**: (Optional) String ID to track usage and costs per unique end-user.
+- **`privacy`**: (Optional) Set to `True` to disable payload logging and session replays for this request. Metadata (latency, tokens) will still be recorded for observability.
+
 ### Supported Models
 
 The SDK provides a `TokenBeeModel` enum with popular models:

{tokenbee_sdk-1.2.2 → tokenbee_sdk-1.3.0}/README.md

@@ -55,20 +55,26 @@ TokenBee is a **stateless** gateway. We do not store your LLM provider API keys
 
 ### Compression Control
 
-You can specify the compression rate and method per request:
+You can specify the compression rate and method per request. TokenBee uses an intelligent semantic engine to reduce token usage while preserving meaning.
 
 ```python
 response = client.send(
     model=TokenBeeModel.ANTHROPIC_CLAUDE_3_5_SONNET,
     input={
         "messages": [...],
-        "compression": "on",
-        "rate": CompressionRate.HIGH,
+        "compression": "auto",      # "auto" (default), "on", or "off"
+        "rate": CompressionRate.HIGH, # MEDIUM (0.5), HIGH (0.33), etc.
         "privacy": True
     }
 )
 ```
 
+- **`compression`**: Set to `"auto"` to let TokenBee decide when to compress, or `"off"` to bypass the compression engine entirely for high-precision tasks.
+- **`rate`**: Controls the aggressiveness of compression. `HIGH` aims for ~67% token reduction.
+- **`sessionId`**: (Optional) String ID to group multiple requests into a single replayable session in the dashboard.
+- **`userId`**: (Optional) String ID to track usage and costs per unique end-user.
+- **`privacy`**: (Optional) Set to `True` to disable payload logging and session replays for this request. Metadata (latency, tokens) will still be recorded for observability.
+
 ### Supported Models
 
 The SDK provides a `TokenBeeModel` enum with popular models:

{tokenbee_sdk-1.2.2 → tokenbee_sdk-1.3.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "tokenbee-sdk"
-version = "1.2.2"
+version = "1.3.0"
 authors = [
   { name="TokenBee Inc.", email="founders@tokenbee.io" },
 ]

{tokenbee_sdk-1.2.2 → tokenbee_sdk-1.3.0}/setup.py

@@ -6,7 +6,7 @@ long_description = (this_directory / "README.md").read_text()
 
 setup(
     name="tokenbee-sdk",
-    version="1.2.2",
+    version="1.3.0",
     packages=find_packages(),
     install_requires=["httpx>=0.23.0"],
     author="TokenBee Inc.",

{tokenbee_sdk-1.2.2 → tokenbee_sdk-1.3.0}/tokenbee/__init__.py

@@ -8,6 +8,17 @@ class CompressionRate(str, Enum):
     HIGH = "0.33"
     EXTREME = "0.2"
 
+class CompressionStrategy(str, Enum):
+    HIVE = "hive_v1"
+    SMART = "smart_v1"
+
+class TokenBeeContext(str, Enum):
+    AUTO = "auto"
+    CONVERSATION = "conversation"
+    DOCUMENT = "document"
+    AGENT = "agent"
+    CODE = "code"
+
 class TokenBeeModel(str, Enum):
     # OpenAI
     OPENAI_GPT_4_5 = "openai/gpt-4.5-preview"
@@ -60,6 +71,8 @@ class TokenBee:
         llm_key: str,
         compression: str = "auto", 
         rate: str = CompressionRate.MEDIUM, 
+        strategy: str = CompressionStrategy.SMART,
+        context: str = TokenBeeContext.AUTO,
         model: str = "", 
         provider: str = "",
         privacy: bool = False
@@ -74,6 +87,8 @@ class TokenBee:
             "X-LLM-Key": llm_key,
             "X-TokenBee-Compression": compression,
             "X-TokenBee-Rate": rate,
+            "X-TokenBee-Strategy": strategy,
+            "X-TokenBee-Context": context,
             "X-TokenBee-Privacy": str(privacy).lower()
         }
         if model:
@@ -96,14 +111,26 @@ class TokenBee:
             headers["X-TokenBee-Compression"] = str(input["compression"])
         if "rate" in input:
             headers["X-TokenBee-Rate"] = str(input["rate"])
+        if "strategy" in input:
+            headers["X-TokenBee-Strategy"] = str(input["strategy"])
+        if "context" in input:
+            headers["X-TokenBee-Context"] = str(input["context"])
         if "privacy" in input:
             headers["X-TokenBee-Privacy"] = str(input["privacy"]).lower()
+        if "sessionId" in input:
+            headers["X-TB-Session-Id"] = str(input["sessionId"])
+        if "userId" in input:
+            headers["X-TB-User-Id"] = str(input["userId"])
 
         payload = input.copy()
         payload["model"] = model_name
         payload.pop("compression", None)
         payload.pop("rate", None)
+        payload.pop("strategy", None)
+        payload.pop("context", None)
         payload.pop("privacy", None)
+        payload.pop("sessionId", None)
+        payload.pop("userId", None)
 
         response = self.client.post("/chat/completions", json=payload, headers=headers)
         response.raise_for_status()

{tokenbee_sdk-1.2.2 → tokenbee_sdk-1.3.0}/tokenbee_sdk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tokenbee-sdk
-Version: 1.2.2
+Version: 1.3.0
 Summary: Official Python SDK for TokenBee LLM inference gateway and observability.
 Author: TokenBee Inc.
 Author-email: "TokenBee Inc." <founders@tokenbee.io>
@@ -73,20 +73,26 @@ TokenBee is a **stateless** gateway. We do not store your LLM provider API keys
 
 ### Compression Control
 
-You can specify the compression rate and method per request:
+You can specify the compression rate and method per request. TokenBee uses an intelligent semantic engine to reduce token usage while preserving meaning.
 
 ```python
 response = client.send(
     model=TokenBeeModel.ANTHROPIC_CLAUDE_3_5_SONNET,
     input={
         "messages": [...],
-        "compression": "on",
-        "rate": CompressionRate.HIGH,
+        "compression": "auto",      # "auto" (default), "on", or "off"
+        "rate": CompressionRate.HIGH, # MEDIUM (0.5), HIGH (0.33), etc.
         "privacy": True
     }
 )
 ```
 
+- **`compression`**: Set to `"auto"` to let TokenBee decide when to compress, or `"off"` to bypass the compression engine entirely for high-precision tasks.
+- **`rate`**: Controls the aggressiveness of compression. `HIGH` aims for ~67% token reduction.
+- **`sessionId`**: (Optional) String ID to group multiple requests into a single replayable session in the dashboard.
+- **`userId`**: (Optional) String ID to track usage and costs per unique end-user.
+- **`privacy`**: (Optional) Set to `True` to disable payload logging and session replays for this request. Metadata (latency, tokens) will still be recorded for observability.
+
 ### Supported Models
 
 The SDK provides a `TokenBeeModel` enum with popular models: