llm_batch_helper 0.1.3__py3-none-any.whl → 0.1.5__py3-none-any.whl

llm_batch_helper/__init__.py

@@ -3,7 +3,7 @@ from .config import LLMConfig
 from .input_handlers import get_prompts, read_prompt_files, read_prompt_list
 from .providers import process_prompts_batch
 
-__version__ = "0.1.1"
+__version__ = "0.1.5"
 
 __all__ = [
     "LLMCache",

llm_batch_helper/providers.py

@@ -60,6 +60,65 @@ async def _get_openai_response_direct(
             "usage_details": usage_details,
         }
 
+
+@retry(
+    stop=stop_after_attempt(5),
+    wait=wait_exponential(multiplier=1, min=4, max=60),
+    retry=retry_if_exception_type(
+        (
+            ConnectionError,
+            TimeoutError,
+            httpx.HTTPStatusError,
+            httpx.RequestError,
+        )
+    ),
+    reraise=True,
+)
+async def _get_together_response_direct(
+    prompt: str, config: LLMConfig
+) -> Dict[str, Union[str, Dict]]:
+    api_key = os.environ.get("TOGETHER_API_KEY")
+    if not api_key:
+        raise ValueError("TOGETHER_API_KEY environment variable not set")
+
+    async with httpx.AsyncClient(timeout=1000.0) as client:
+        messages = [
+            {"role": "system", "content": config.system_instruction},
+            {"role": "user", "content": prompt},
+        ]
+
+        headers = {
+            "Authorization": f"Bearer {api_key}",
+            "Content-Type": "application/json",
+        }
+
+        payload = {
+            "model": config.model_name,
+            "messages": messages,
+            "temperature": config.temperature,
+            "max_tokens": config.max_completion_tokens,
+        }
+
+        response = await client.post(
+            "https://api.together.xyz/chat/completions",
+            json=payload,
+            headers=headers,
+        )
+        response.raise_for_status()
+        
+        response_data = response.json()
+        usage = response_data.get("usage", {})
+        usage_details = {
+            "prompt_token_count": usage.get("prompt_tokens", 0),
+            "completion_token_count": usage.get("completion_tokens", 0),
+            "total_token_count": usage.get("total_tokens", 0),
+        }
+        
+        return {
+            "response_text": response_data["choices"][0]["message"]["content"],
+            "usage_details": usage_details,
+        }
+
 async def get_llm_response_with_internal_retry(
     prompt_id: str,
     prompt: str,
@@ -77,6 +136,8 @@ async def get_llm_response_with_internal_retry(
     try:
         if provider.lower() == "openai":
             response = await _get_openai_response_direct(prompt, config)
+        elif provider.lower() == "together":
+            response = await _get_together_response_direct(prompt, config)
         else:
             raise ValueError(f"Unsupported provider: {provider}")
 
@@ -107,7 +168,7 @@ async def process_prompts_batch(
         prompts: Optional list of prompts in any supported format (string, tuple, or dict)
         input_dir: Optional path to directory containing prompt files
         config: LLM configuration
-        provider: LLM provider to use ("openai" or "gemini")
+        provider: LLM provider to use ("openai", "together", or "gemini")
         desc: Description for progress bar
         cache_dir: Optional directory for caching responses
         force: If True, force regeneration even if cached response exists

{llm_batch_helper-0.1.3.dist-info → llm_batch_helper-0.1.5.dist-info}/METADATA

@@ -1,9 +1,9 @@
 Metadata-Version: 2.3
 Name: llm_batch_helper
-Version: 0.1.3
+Version: 0.1.5
 Summary: A Python package that enables batch submission of prompts to LLM APIs, with built-in async capabilities and response caching.
 License: MIT
-Keywords: llm,openai,batch,async,ai,nlp,api
+Keywords: llm,openai,together,batch,async,ai,nlp,api
 Author: Tianyi Peng
 Author-email: tianyipeng95@gmail.com
 Requires-Python: >=3.11,<4.0
@@ -17,7 +17,7 @@ Classifier: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
-Requires-Dist: httpx (>=0.24.0,<0.25.0)
+Requires-Dist: httpx (>=0.24.0,<2.0.0)
 Requires-Dist: openai (>=1.0.0,<2.0.0)
 Requires-Dist: python-dotenv (>=1.0.0,<2.0.0)
 Requires-Dist: tenacity (>=8.0.0,<9.0.0)
@@ -35,7 +35,7 @@ A Python package that enables batch submission of prompts to LLM APIs, with buil
 - **Async Processing**: Submit multiple prompts concurrently for faster processing
 - **Response Caching**: Automatically cache responses to avoid redundant API calls
 - **Multiple Input Formats**: Support for both file-based and list-based prompts
-- **Provider Support**: Works with OpenAI API
+- **Provider Support**: Works with OpenAI and Together.ai APIs
 - **Retry Logic**: Built-in retry mechanism with exponential backoff
 - **Verification Callbacks**: Custom verification for response quality
 - **Progress Tracking**: Real-time progress bars for batch operations
@@ -70,6 +70,9 @@ poetry shell
 ```bash
 # For OpenAI
 export OPENAI_API_KEY="your-openai-api-key"
+
+# For Together.ai
+export TOGETHER_API_KEY="your-together-api-key"
 ```
 
 ### 2. Interactive Tutorial (Recommended)
@@ -89,7 +92,7 @@ async def main():
     config = LLMConfig(
         model_name="gpt-4o-mini",
         temperature=0.7,
-        max_tokens=100,
+        max_completion_tokens=100,  # or use max_tokens for backward compatibility
         max_concurrent_requests=30 # number of concurrent requests with asyncIO
     )
     
@@ -127,7 +130,7 @@ async def process_files():
     config = LLMConfig(
         model_name="gpt-4o-mini",
         temperature=0.7,
-        max_tokens=200
+        max_completion_tokens=200
     )
     
     # Process all .txt files in a directory
@@ -183,7 +186,8 @@ Configuration class for LLM requests.
 LLMConfig(
     model_name: str,
     temperature: float = 0.7,
-    max_tokens: Optional[int] = None,
+    max_completion_tokens: Optional[int] = None,  # Preferred parameter
+    max_tokens: Optional[int] = None,  # Deprecated, kept for backward compatibility
     system_instruction: Optional[str] = None,
     max_retries: int = 10,
     max_concurrent_requests: int = 5,
@@ -199,7 +203,7 @@ Main function for batch processing of prompts.
 ```python
 async def process_prompts_batch(
     config: LLMConfig,
-    provider: str,  # "openai"
+    provider: str,  # "openai" or "together"
     prompts: Optional[List[str]] = None,
     input_dir: Optional[str] = None,
     cache_dir: str = "llm_cache",
@@ -256,6 +260,23 @@ llm_batch_helper/
 - gpt-4
 - gpt-3.5-turbo
 
+### Together.ai
+- meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo
+- meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo
+- mistralai/Mixtral-8x7B-Instruct-v0.1
+- And many other open-source models
+
+## Documentation
+
+📖 **[Complete Documentation](https://llm-batch-helper.readthedocs.io/)** - Comprehensive docs on Read the Docs
+
+### Quick Links:
+- [Quick Start Guide](https://llm-batch-helper.readthedocs.io/en/latest/quickstart.html) - Get started quickly
+- [API Reference](https://llm-batch-helper.readthedocs.io/en/latest/api.html) - Complete API documentation  
+- [Examples](https://llm-batch-helper.readthedocs.io/en/latest/examples.html) - Practical usage examples
+- [Tutorials](https://llm-batch-helper.readthedocs.io/en/latest/tutorials.html) - Step-by-step tutorials
+- [Provider Guide](https://llm-batch-helper.readthedocs.io/en/latest/providers.html) - OpenAI & Together.ai setup
+
 ## Contributing
 
 1. Fork the repository
@@ -271,6 +292,12 @@ This project is licensed under the MIT License - see the [LICENSE](LICENSE) file
 
 ## Changelog
 
+### v0.1.5
+- Added Together.ai provider support
+- Support for open-source models (Llama, Mixtral, etc.)
+- Enhanced documentation with Read the Docs
+- Updated examples and tutorials
+
 ### v0.1.0
 - Initial release
 - Support for OpenAI API

llm_batch_helper-0.1.5.dist-info/RECORD

@@ -0,0 +1,10 @@
+llm_batch_helper/__init__.py,sha256=POB4Fodeltq96NbiaLh7YSEPwEu50Giz46V2qyVZZoY,348
+llm_batch_helper/cache.py,sha256=QUODQ1tPCvFThO3yvVOTcorcOrmN2dP5HLF1Y2O1bTQ,1276
+llm_batch_helper/config.py,sha256=WcZKTD-Mtocsx1plS9x3hh6MstVmyxD-tyidGUatkPY,1327
+llm_batch_helper/exceptions.py,sha256=59_f3jINUhKFble6HTp8pmtLSFE2MYLHWGclwaQKs28,296
+llm_batch_helper/input_handlers.py,sha256=IadA732F1Rw0zcBok5hjZr32RUm8eTUOpvLsRuMvaE4,2877
+llm_batch_helper/providers.py,sha256=aV6IbGfRqFoxCQ90yd3UsCqmyeOBMRC9YW8VVq6ghq8,10258
+llm_batch_helper-0.1.5.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+llm_batch_helper-0.1.5.dist-info/METADATA,sha256=58ray3o9P37IjYcgXfPa_SS5YnQhz3M212zrxa0e3L0,8882
+llm_batch_helper-0.1.5.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+llm_batch_helper-0.1.5.dist-info/RECORD,,

llm_batch_helper-0.1.3.dist-info/RECORD

@@ -1,10 +0,0 @@
-llm_batch_helper/__init__.py,sha256=WuZSgItWxi-uoND-X-tOmYaxJaxPXAUa9AXZM58bNuc,348
-llm_batch_helper/cache.py,sha256=QUODQ1tPCvFThO3yvVOTcorcOrmN2dP5HLF1Y2O1bTQ,1276
-llm_batch_helper/config.py,sha256=WcZKTD-Mtocsx1plS9x3hh6MstVmyxD-tyidGUatkPY,1327
-llm_batch_helper/exceptions.py,sha256=59_f3jINUhKFble6HTp8pmtLSFE2MYLHWGclwaQKs28,296
-llm_batch_helper/input_handlers.py,sha256=IadA732F1Rw0zcBok5hjZr32RUm8eTUOpvLsRuMvaE4,2877
-llm_batch_helper/providers.py,sha256=ccvDqn136fMWPWaqfSFUToeOJFFhSI86YatLoH1wrQs,8332
-llm_batch_helper-0.1.3.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-llm_batch_helper-0.1.3.dist-info/METADATA,sha256=oJgWbwZQpPicAv63GX83kF1I6kNxXqPKQBIgCBqHgoc,7525
-llm_batch_helper-0.1.3.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-llm_batch_helper-0.1.3.dist-info/RECORD,,