ScreenerBot
DocsAssistantAI Providers

Assistant Providers Reference

Complete reference for all 10+ supported Assistant providers in ScreenerBot. Learn about pricing, free tiers, recommended models, and setup instructions for each provider.

Provider Comparison

ProviderFree TierRecommended ModelBest For
Groq
30 RPMllama-3.1-70b-versatileUltra-fast real-time trading
DeepSeek
500K/daydeepseek-chatHigh-volume screening
Gemini
1500/daygemini-1.5-flashLarge context analysis
Ollama
Unlimitedllama3.2Local/private, zero cost
OpenRouter
Free creditsVariesUnified API, many models
OpenAI
Trial onlygpt-4o-miniHighest accuracy
Anthropic
Trial onlyclaude-3-5-sonnetBest reasoning
Mistral
Free tiermistral-large-latestFast EU provider
Together AI
$1 creditLlama-3-70b200+ models available
Fireworks
Trialllama-v3-70bLow latency inference

Provider Setup Guides

Groq

Details:

  • Free Tier: 30 requests/minute
  • Speed: Fastest (~100ms latency)
  • Models: Llama, Mixtral, Gemma
  • Website: console.groq.com

Recommended Models:

  • llama-3.1-70b-versatile - Best overall
  • llama-3.1-8b-instant - Fastest
  • mixtral-8x7b - Good balance

Why choose Groq: Perfect for real-time trading where milliseconds matter. Free tier is generous and sufficient for most use cases. Best choice for getting started.

OpenAI

Details:

  • Free Tier: $5 trial credits (limited time)
  • Pricing: Pay-per-use after trial
  • Models: GPT-4o, GPT-4o-mini, GPT-3.5
  • Website: platform.openai.com

Recommended Models:

  • gpt-4o-mini - Best value, accurate
  • gpt-4o - Highest quality
  • gpt-3.5-turbo - Cheapest option

Why choose OpenAI: Industry-leading accuracy for complex analysis. GPT-4o-mini offers the best balance of cost and performance for token analysis tasks.

DeepSeek

Details:

  • Free Tier: 500,000 requests/day
  • Pricing: Extremely cheap after free tier
  • Models: DeepSeek-Chat, DeepSeek-Coder
  • Website: platform.deepseek.com

Recommended Models:

  • deepseek-chat - Main model
  • deepseek-reasoner - Enhanced reasoning

Why choose DeepSeek: Unbeatable free tier for high-volume screening. Excellent for analyzing hundreds of tokens daily without worrying about costs.

Anthropic (Claude)

Details:

  • Free Tier: $5 trial credits
  • Pricing: Pay-per-use after trial
  • Models: Claude 3.5, Claude 3
  • Website: console.anthropic.com

Recommended Models:

  • claude-3-5-sonnet - Best reasoning
  • claude-3-haiku - Fast & cheap
  • claude-3-opus - Most capable

Why choose Anthropic: Superior reasoning abilities for complex risk assessment. Claude excels at identifying subtle scam indicators and providing nuanced analysis.

Google Gemini

Details:

  • Free Tier: 1,500 requests/day
  • Context: Up to 2M tokens
  • Models: Gemini 1.5 Pro/Flash
  • Website: aistudio.google.com

Recommended Models:

  • gemini-1.5-flash - Fast, balanced
  • gemini-1.5-pro - Most capable

Why choose Gemini: Massive context window allows analyzing tokens with extensive historical data. Good free tier and competitive pricing for paid usage.

Ollama (Local)

Details:

  • Free Tier: Unlimited (runs on your hardware)
  • Privacy: 100% local, no API calls
  • Requirements: 8GB+ RAM recommended
  • Website: ollama.com

Recommended Models:

  • llama3.2 - Good balance
  • mistral - Fast inference
  • qwen2.5 - High quality

Why choose Ollama: Perfect for privacy-conscious traders or those wanting zero API costs. Requires decent hardware but offers unlimited usage without rate limits.

Installation:

# Install Ollama
curl -fsSL https://ollama.com/install.sh | sh

# Pull a model
ollama pull llama3.2

# Verify it's running
ollama list

OpenRouter

Details:

  • Free Tier: Free credits for new users
  • Models: Access to 100+ models via one API
  • Pricing: Varies by model
  • Website: openrouter.ai

Popular Models:

  • anthropic/claude-3.5-sonnet
  • openai/gpt-4o-mini
  • meta-llama/llama-3.1-70b

Why choose OpenRouter: Single API for multiple providers. Easy switching between models without reconfiguring. Great for testing different models quickly.

Mistral AI

Details:

  • Free Tier: Available with rate limits
  • Region: European provider
  • Models: Mistral Large, Medium, Small
  • Website: console.mistral.ai

Recommended Models:

  • mistral-large-latest - Best quality
  • mistral-small-latest - Fast & cheap

Why choose Mistral: Strong European alternative with good performance and pricing. Fast inference and multilingual capabilities.

Together AI

Details:

  • Free Tier: $1 free credits
  • Models: 200+ open-source models
  • Pricing: Competitive pay-per-use
  • Website: api.together.xyz

Recommended Models:

  • meta-llama/Llama-3-70b
  • mistralai/Mixtral-8x7B
  • Qwen/Qwen2-72B

Why choose Together AI: Massive model selection including cutting-edge open-source models. Great for experimenting with different architectures.

Fireworks AI

Details:

  • Free Tier: Trial credits available
  • Speed: Optimized for low latency
  • Models: Llama, Mixtral, and more
  • Website: fireworks.ai

Recommended Models:

  • llama-v3-70b-instruct
  • mixtral-8x7b-instruct

Why choose Fireworks: Production-grade inference infrastructure with consistently low latency. Ideal for high-frequency trading scenarios.

Choosing the Right Provider

For Getting Started

Best free options for testing and learning:

  • Groq - Start here for speed
  • DeepSeek - High volume testing
  • Ollama - Local development

For Production Trading

Best paid options for live trading:

  • OpenAI GPT-4o-mini - Best accuracy
  • Anthropic Claude - Best reasoning
  • Groq - Best speed (paid tier)

Next Steps