Recommended Models

ServerAssistantAI supports a wide range of language models from different providers, including premium and free options, allowing server owners to choose the model that best fits their needs. Here's what we recommend, listed from most recommended to least.

Embedding Models

Embedding models are used to convert text data into numerical representations called embeddings. These embeddings capture the semantic meaning and relationships between different pieces of text. When the documents/ directory is updated, the content is sent to the embedding API. The resulting embeddings are saved to the cache/ directory, allowing the AI to find relevant context efficiently without reprocessing or making new API requests for each query.

Provider	Embedding Model	Pricing	Description
Google-AIStudio	text-embedding-004	Free & Paid	Google's state-of-the-art embeddings for semantic search and classification tasks.
OpenAI	text-embedding-3-large	Paid	OpenAI's large embedding model for text similarity and retrieval tasks.
Cohere	embed-english-v3.0	Free & Paid	English embedding model great for noisy data, enabling better retrievals for RAG.
Cohere	embed-multilingual-v3.0	Free & Paid	Multilingual embedding model great for noisy data, enabling better retrievals for RAG.
OpenAI	text-embedding-3-small	Paid	OpenAI's small embedding model for text similarity and retrieval tasks.

ServerAssistantAI only sends information to the embedding API when changes are made to the documents/ directory. If no changes are detected, the plugin will use the previously cached embeddings to reduce API calls.

Large Language Models

Large Language Models (LLMs) are powerful AI models that can understand and generate human-like text based on the input they receive. In ServerAssistantAI, when a user asks a question, the system retrieves relevant cached context from the embedding API results. This context, along with the user's question, is sent to the LLM to generate accurate and context-aware responses.

Provider	Large Language Model	Pricing	Description
OpenAI	gpt-4o	Paid	OpenAI's most advanced model. Same intelligence as GPT-4 Turbo but 2x faster and 50% cheaper.
Github-Models	gpt-4o	Free & Paid	OpenAI's most advanced model. Same intelligence as GPT-4 Turbo but 2x faster and 50% cheaper.
Anthropic	claude-opus-4.1-20250805	Paid	Anthropic's upgraded model with enhanced agentic tasks, real-world coding, and reasoning capabilities.
Anthropic	claude-opus-4-20250514	Paid	Anthropic's most intelligent model with hybrid reasoning and 200K context window, excelling at coding and working autonomously for hours.
Google-AIStudio	gemini-2.5-flash	Free & Paid	Google's fast thinking model optimized for speed and cost-efficiency with controllable thinking budgets for high-volume tasks.
OpenAI	gpt-4.1	Paid	OpenAI's enhanced model with major coding improvements and 1 million token context window.
Github-Models	gpt-4.1	Free & Paid	OpenAI's enhanced model with major coding improvements and 1 million token context window.
Groq	deepseek-r1-distill-llama-70b	Free & Paid	DeepSeek's efficient 70B reasoning model distilled from larger models.
Anthropic	claude-sonnet-4-20250514	Paid	Anthropic's high-performance model with 200K context window, achieving 72.7% on SWE-bench with superior instruction following.
OpenAI	gpt-4.1-mini	Paid	OpenAI's efficient model competitive with GPT-4o but 50% faster and 83% cheaper with 1 million token context window.
Github-Models	gpt-4.1-mini	Free & Paid	OpenAI's efficient model competitive with GPT-4o but 50% faster and 83% cheaper with 1 million token context window.
Google-AIStudio	gemini-2.0-flash-001	Free & Paid	Google's next-generation flash model with native tool use, superior speed, and 1M token context window for multimodal applications.
Google-AIStudio	gemma-3-27b-it	Free	Google's open-source 27B parameter model optimized for instruction following and conversational AI with efficient performance.
Google-AIStudio	gemini-1.5-pro	Free & Paid	Gemini 1.5 Pro offers long-context understanding, with a context window of up to 1 million tokens.
Groq	qwen/qwen3-32b	Free & Paid	Alibaba's hybrid model switching between thinking and fast modes, supporting 119 languages and competing with top proprietary models.
Cohere	command-a-03-2025	Free & Paid	Cohere's most performant 111B model with 256K context, delivering 150% higher throughput.
Anthropic	claude-3-sonnet-20240307	Paid	Claude 3.5 Sonnet delivers enhanced intelligence and speed, ideal for advanced tasks.
OpenAI	gpt-4o-mini	Paid	OpenAI's most cost-efficient small model that’s smarter and cheaper than mostly all other paid models.
Github-Models	gpt-4o-mini	Free & Paid	OpenAI's most cost-efficient small model that’s smarter and cheaper than mostly all other paid models.
Groq	llama-3.1-70b-versatile	Free & Paid	Meta's Llama 3.1 70B instruction-tuned model outperforms all open-source chat models and even closed-source models and using Groq, the inference speed is 250+ Tokens per second!
OpenAI	gpt-4-turbo	Paid	OpenAI's GPT-4-Turbo model with 128K context, newer knowledge and more powerful than GPT-4.
Google-AIStudio	gemini-1.5-flash	Free & Paid	Gemini 1.5 Flash has higher rate limits than Gemini 1.5 Pro and is Google's fastest, most cost-efficient model.
Anthropic	claude-3-opus-20240307	Paid	Anthropic's most intelligent model with great performance on highly complex tasks.
Cohere	command-r-plus-08-2024	Free & Paid	Updated version of Command R+, with advanced RAG model with 128k context, multilingual support, and tool use capabilities.
Groq	llama3-70b-8192	Free & Paid	Meta's Llama 3 70B instruction-tuned model which outperforms almost all open-source chat models and using Groq, the inference speed is 300+ Tokens per second!
Cohere	command-r-plus	Free & Paid	Advanced RAG model with 128k context, multilingual support, and tool use capabilities.
Anthropic	claude-3-haiku-20240307	Paid	Claude 3 Haiku is Anthropic's fastest model for near-instant responsiveness.
Cohere	command-r	Free & Paid	Scalable generative model for RAG and Tool Use in enterprise applications.
HuggingFace	meta-llama/Meta-Llama-3-8B-Instruct	Free	Meta's Llama 3 8B instruction-tuned model and outperforming many open-source chat models.
HuggingFace	01-ai/Yi-1.5-34B-Chat	Free	01.AI's Yi-1.5 model, offering strong performance in coding, math, reasoning, and instruction-following.
HuggingFace	mistralai/Mixtral-8x7B-Instruct-v0.1	Free	Mistral AI's 8x7B instruction-following model, version 0.1.
OpenAI	gpt-3.5-turbo	Paid	OpenAI's GPT-3.5 model with a larger context size of 16K tokens.

Please note that each model may have specific requirements or considerations for optimal performance. When selecting a model, consider factors such as:

• System prompt configuration: Different models may require adjustments to the system prompt to achieve the best results. Check the documentation provided by the model's creators for tips on prompt engineering and configuration.

• Open-source free LLMs have a higher chance of hallucinating compared to paid models.