Picture by Writer | ideogram.ai# Introduction
With the surge of huge language fashions (LLMs) lately, many LLM-powered purposes are rising. LLM implementation has launched options that have been beforehand non-existent.
As time goes on, many LLM fashions and merchandise have grow to be accessible, every with its professionals and cons. Sadly, there’s nonetheless no customary solution to entry all these fashions, as every firm can develop its personal framework. That’s the reason having an open-source software similar to LiteLLM is helpful once you want standardized entry to your LLM apps with none further value.
On this article, we’ll discover why LiteLLM is useful for constructing LLM purposes.
Let’s get into it.
# Profit 1: Unified Entry
LiteLLM’s greatest benefit is its compatibility with totally different mannequin suppliers. The software helps over 100 totally different LLM companies by way of standardized interfaces, permitting us to entry them whatever the mannequin supplier we use. It’s particularly helpful in case your purposes make the most of a number of totally different fashions that have to work interchangeably.
A couple of examples of the foremost mannequin suppliers that LiteLLM helps embody:
- OpenAI and Azure OpenAI, like GPT-4.
- Anthropic, like Claude.
- AWS Bedrock & SageMaker, supporting fashions like Amazon Titan and Claude.
- Google Vertex AI, like Gemini.
- Hugging Face Hub and Ollama for open-source fashions like LLaMA and Mistral.
The standardized format follows OpenAI’s framework, utilizing its chat/completions schema. Which means that we will change fashions simply without having to know the unique mannequin supplier’s schema.
For instance, right here is the Python code to make use of Google’s Gemini mannequin with LiteLLM.
from litellm import completion
immediate = "YOUR-PROMPT-FOR-LITELLM"
api_key = "YOUR-API-KEY-FOR-LLM"
response = completion(
mannequin="gemini/gemini-1.5-flash-latest",
messages=[{"content": prompt, "role": "user"}],
api_key=api_key)
response['choices'][0]['message']['content']
You solely have to get hold of the mannequin title and the respective API keys from the mannequin supplier to entry them. This flexibility makes LiteLLM supreme for purposes that use a number of fashions or for performing mannequin comparisons.
# Profit 2: Price Monitoring and Optimization
When working with LLM purposes, it is very important monitor token utilization and spending for every mannequin you implement and throughout all built-in suppliers, particularly in real-time eventualities.
LiteLLM permits customers to keep up an in depth log of mannequin API name utilization, offering all the required data to manage prices successfully. For instance, the `completion` name above could have details about the token utilization, as proven beneath.
utilization=Utilization(completion_tokens=10, prompt_tokens=8, total_tokens=18, completion_tokens_details=None, prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None, text_tokens=8, image_tokens=None))
Accessing the response’s hidden parameters may also present extra detailed data, together with the price.
With the output much like beneath:
{'custom_llm_provider': 'gemini',
'region_name': None,
'vertex_ai_grounding_metadata': [],
'vertex_ai_url_context_metadata': [],
'vertex_ai_safety_results': [],
'vertex_ai_citation_metadata': [],
'optional_params': {},
'litellm_call_id': '558e4b42-95c3-46de-beb7-9086d6a954c1',
'api_base': 'https://generativelanguage.googleapis.com/v1beta/fashions/gemini-1.5-flash-latest:generateContent',
'model_id': None,
'response_cost': 4.8e-06,
'additional_headers': {},
'litellm_model_name': 'gemini/gemini-1.5-flash-latest'}
There’s numerous data, however an important piece is `response_cost`, because it estimates the precise cost you’ll incur throughout that decision, though it may nonetheless be offset if the mannequin supplier affords free entry. Customers may also outline customized pricing for fashions (per token or per second) to calculate prices precisely.
A extra superior cost-tracking implementation may also enable customers to set a spending finances and restrict, whereas additionally connecting the LiteLLM value utilization data to an analytics dashboard to extra simply mixture data. It is also potential to supply customized label tags to assist attribute prices to sure utilization or departments.
By offering detailed value utilization knowledge, LiteLLM helps customers and organizations optimize their LLM software prices and finances extra successfully.
# Profit 3: Ease of Deployment
LiteLLM is designed for straightforward deployment, whether or not you utilize it for native growth or a manufacturing surroundings. With modest sources required for Python library set up, we will run LiteLLM on our native laptop computer or host it in a containerized deployment with Docker with no want for advanced further configuration.
Talking of configuration, we will arrange LiteLLM extra effectively utilizing a YAML config file to checklist all the required data, such because the mannequin title, API keys, and any important customized settings to your LLM Apps. It’s also possible to use a backend database similar to SQLite or PostgreSQL to retailer its state.
For knowledge privateness, you’re liable for your individual privateness as a person deploying LiteLLM your self, however this strategy is safer because the knowledge by no means leaves your managed surroundings besides when despatched to the LLM suppliers. One characteristic LiteLLM supplies for enterprise customers is Single Signal-On (SSO), role-based entry management, and audit logs in case your software wants a safer surroundings.
General, LiteLLM supplies versatile deployment choices and configuration whereas preserving the info safe.
# Profit 4: Resilience Options
Resilience is essential when constructing LLM Apps, as we would like our software to stay operational even within the face of sudden points. To advertise resilience, LiteLLM supplies many options which can be helpful in software growth.
One characteristic that LiteLLM has is built-in caching, the place customers can cache LLM prompts and responses in order that equivalent requests do not incur repeated prices or latency. It’s a helpful characteristic if our software ceaselessly receives the identical queries. The caching system is versatile, supporting each in-memory and distant caching, similar to with a vector database.
One other characteristic of LiteLLM is automated retries, permitting customers to configure a mechanism when requests fail attributable to errors like timeouts or rate-limit errors to robotically retry the request. It’s additionally potential to arrange further fallback mechanisms, similar to utilizing one other mannequin if the request has already hit the retry restrict.
Lastly, we will set price limiting for outlined requests per minute (RPM) or tokens per minute (TPM) to restrict the utilization degree. It’s a good way to cap particular mannequin integrations to forestall failures and respect software infrastructure necessities.
# Conclusion
Within the period of LLM product progress, it has grow to be a lot simpler to construct LLM purposes. Nonetheless, with so many mannequin suppliers on the market, it turns into exhausting to determine a normal for LLM implementation, particularly within the case of multi-model system architectures. This is the reason LiteLLM will help us construct LLM Apps effectively.
I hope this has helped!
Cornellius Yudha Wijaya is a knowledge science assistant supervisor and knowledge author. Whereas working full-time at Allianz Indonesia, he likes to share Python and knowledge ideas by way of social media and writing media. Cornellius writes on quite a lot of AI and machine studying subjects.
