research additional LLM providers

[cobycloud]

• [ ] xAI's grok
• [ ] AWS

[cobycloud]

• [x] ai21
• [ ] ailayer
• [x] aimlapi
• [x] anthropic
• [ ] anyscale
• [ ] azureai
• [ ] cloudflareai
• [x] cohere
• [ ] corcel
• [ ] deepinfra
• [x] deepseek
• [ ] fireworksai
• [x] forefront
• [ ] friendliai
• [x] gemini
• [x] gooseai
• [x] groq
• [ ] huggingface
• [ ] hyperbeeai
• [x] lamini
• [ ] llamacpp
• [x] mistralai
• [ ] monsterapi
• [ ] neetsai
• [ ] novitaai
• [ ] nvidia
• [ ] octoai
• [ ] ollama
• [x] openai
• [x] perplexity
• [ ] rekaai
• [ ] replicate
• [x] shuttleai
• [ ] siliconflow
• [ ] thebai
• [ ] togetherai
• [ ] voyage
• [ ] watsonxai
• [ ] writer
• [ ] zhipuai

[cobycloud]

assigned ai21 to faizan2700

[cobycloud]

assigned deepseek to faizan2700

[cobycloud]

replicate.com is very diverse and will require many subclasses to handle the different parameters

[cobycloud]

• [ ] bing
• [ ] figgs.ai
• [ ] cinematika
• [ ] stabilityai
• [ ] seco
• [ ] LCM
• [ ] deepfloyd
• [ ] material
• [ ] elevenlab
• [ ] speechify
• [ ] duckduck

[cobycloud]

• [ ] blackbox
• [ ] microsoft
• [ ] cognitivecomputations

[cobycloud]

we need to:

1. create a list of all of the LLM provider options

2. iterate through that list gathering the documentation links for their home page, platform registration, models supported, pricing, and how to use their API via python.

3. create a new issue for each chat completion model discovered, detailing the file path of the new component and test file, default model, allowed models, and test case ENV variable key names.

4. create a new issue for each tool model discovered, detailing the file path of the new component and test file, default model, allowed models, and test case ENV variable key names.