Model Catalog
| Model | Description | Category | License | Size | Context | Capabilities | Output Formats | Billing Unit |
|---|---|---|---|---|---|---|---|---|
| GPT-OSS 20B | General-purpose model optimized for text generation, automation, and conversational use cases, with support for function calling and structured outputs. | LLM Light + Reasoning | Apache 2.0 | 21.5B | 131K | Function calling, reasoning, streaming | Raw text, JSON Object, JSON Schema | Input/output tokens |
| GPT-OSS 120B | Large-scale model oriented toward complex reasoning tasks, advanced analysis, and agentic workflows. | LLM Reasoning | Apache 2.0 | 120B | 131K | Function calling, reasoning, streaming | Raw text, JSON Object, JSON Schema | Input/output tokens |
| DeepSeek R1 32B | Model focused on advanced reasoning and solving complex multi-step problems. | LLM Reasoning | MIT | 32B | 128K | Reasoning, chain-of-thought, streaming | Raw text | Input/output tokens |
| Llama 3.1 8B Instruct | Lightweight general-purpose model, suitable for chatbots, virtual assistants, and process automation. | LLM Light | Llama 3.1 | 8.3B | 128K | Function calling, multilingual, streaming | Raw text, JSON Object, JSON Schema | Input/output tokens |
| Mistral Nemo Instruct 2407 | European model optimized for enterprise applications, content generation, and conversational assistance. | LLM Light + European | Apache 2.0 | 12.2B | 128K | Function calling, multilingual, streaming | Raw text, JSON Object, JSON Schema | Input/output tokens |
| Qwen 2.5 7B Instruct | Compact and versatile model for chatbots, automation, text generation, and application integration. | LLM Light | Apache 2.0 | 7B | 128K | Function calling, structured output, streaming | Raw text, JSON Object, JSON Schema | Input/output tokens |
| Gemma-4-31B-IT-NVFP4 | Lightweight multimodal model specialized in analyzing complex images and video. | LMM Light + Multimodal | Apache 2.0 | 30.7B | 256K | Function calling, reasoning, multimodal (images + video), streaming | Raw text, JSON Object | Input/output tokens |
| Mistral Small 3.2 24B Instruct | European multimodal model oriented toward enterprise scenarios requiring text and image understanding. | LMM Multimodal + European | Apache 2.0 | 24B | 128K | Function calling, multimodal, streaming | Raw text, JSON Object, JSON Schema | Input/output tokens |
| Qwen 3.5 122B A10B FP8 | Large-scale multimodal model designed for advanced reasoning, AI agents, and complex workflows. | LMM Large + Multimodal | Apache 2.0 | 122B | 262K | Function calling, multimodal, reasoning, agents, streaming | Raw text, JSON Object, JSON Schema | Input/output tokens |
| Qwen3 Coder 30B Instruct | Model specialized in software development, code generation, and developer assistance. | LLM Coding | Apache 2.0 | 30B | 256K | Function calling, coding assistant, streaming | Raw text, JSON Object | Input/output tokens |
| Qwen3 Coder Next | Advanced model for software development and automation of complex coding tasks. | LLM Large + Coding | Apache 2.0 | 80B | 256K | Function calling, coding assistant, streaming | Raw text, JSON Object | Input/output tokens |
| Multilingual E5 Large | Multilingual embedding model for semantic search, similarity search, and Retrieval Augmented Generation (RAG). | Embedding | MIT | 560M | 512 tokens | Embedding retrieval, semantic similarity | Embedding | Input tokens |
| BGE-M3 | Embedding model optimized for advanced retrieval and enterprise semantic search scenarios. | Embedding | MIT | 567M | 8,192 tokens | Dense retrieval, sparse retrieval, multi-vector retrieval | Embedding | Input tokens |
| Qwen3 Guard Gen 8B | Model dedicated to content moderation and safety classification of prompts and outputs. | Guard / Safety | Apache 2.0 | 8B | 32K | Content moderation, safety classification | Raw text | Input/output tokens |
| Qwen Image 2512 | Model for generating images from text prompts. | Image Generation | Apache 2.0 | 20B | N/A | Text-to-Image | PNG / JPEG | Image / call |
| Qwen Image Edit 2511 | Model for editing and transforming images via text prompts and input images. | Image Editing | Apache 2.0 | 20B | N/A | Image Editing, Multi-image Input | PNG / JPEG | Image / call |
AI Error Disclaimer
Content generated by AI may be inaccurate, incomplete, or outdated. It must be reviewed and verified by a qualified person and must not be used as the sole basis for decisions or actions. The user remains responsible for its use and for any resulting consequences.
Data Usage and AI Output Disclaimer
The data provided is not used to train AI models. However, AI-generated outputs may be inaccurate, incomplete, or outdated and must be reviewed and verified before use. The user remains responsible for evaluating and using such outputs.