Models

Model Name	Specialisation & Notes
llama-3.1-8b-instruct	Fast, cost-efficient for general conversation, summarization, and Q&A
llama-3.3-70b-instruct	High-accuracy, large-context, detailed reasoning, enterprise needs
llama3-70b-8192	Big-context, top-tier language, best for long documents
llama3-8b-8192	Quick, lightweight, versatile for broad text use cases
gemma2-9b-it	Instruction tuning, helpful for task-oriented chat
deepseek-v3-0324-671b	High-end, advanced reasoning, suited for research, code, analysis
deepseek-r1-671b	Premium, very large, maximal language competence
deepseek-r1-distill-llama-70b	Efficient but potent, optimized for quick, quality responses
qwen3-32b	Robust, good at multi-turn, diverse languages, instruction following
qwq-32b	Comparable to Qwen3-32b, robust on dialogue and multilingual tasks
gpt-oss-20b	Open-source variant, 20B params, suitable for balanced resource use and mid-scale needs (summary based on model size & naming convention).
gpt-oss-120b	Open-source large-scale, 120B params, suited for research/enterprise scale, handles extended context length and advanced language modeling.

Model Name	Specialisation & Notes
image-gen.pro-1.0	Advanced image generation; high-res outputs, varied visual styles, image editing, creating new image from multiple images

Model Name	Specialisation & Notes
video-gen.pro-1.0	Text-to-video & image-to-video; rich media & configurable durations
video-gen.lite	Image-to-video; rich media & configurable durations