Model List & Capability Comparison
Models & Pricing
Model List & Capability Comparison View all AI models supported by the platform and their capability comparisons
The platform currently supports GPT, Claude Haiku, Claude Opus, Claude Sonnet, and Gemini models. The following is an overview of the primary model families. Other listed models will be connected gradually.
Model Family Representative Models Context Window Key Features Access Status OpenAI GPT GPT-5.5, GPT-5.4, GPT-5.4-mini 1M Flagship performance; supports vision and function calling ✅ Supported Anthropic Claude Claude Opus 4.8, Claude Opus 4.7, Claude Sonnet 4.6, Claude Haiku 4.5 1M Exceptional at long-document comprehension, writing, and coding ✅ Supported Google Gemini Gemini 3.1 Pro 1M Ultra-long context; cutting-edge performance for agents and code ✅ Supported DeepSeek DeepSeek-V4 Pro, DeepSeek-V4 Flash 64K Excellent Chinese comprehension and code generation; very cost-effective Coming Soon Alibaba Qwen Qwen3-235B, Qwen3-72B 128K Optimized for Chinese use cases; enterprise-grade applications Coming Soon Meta Llama Llama 4 Scout, Llama 3.3 128K Open-source models with high value-for-money Coming Soon
Model Context Window Vision Function Calling Reasoning Best For Access Status GPT-5.5 1M ✅ ✅ ⭐⭐⭐⭐⭐ Flagship reasoning, code, professional tasks ✅ Supported GPT-5.4 1M ✅ ✅ ⭐⭐⭐⭐⭐ General chat, code, professional tasks ✅ Supported GPT-5.4-mini 400K ✅ ✅ ⭐⭐⭐⭐ High-frequency, low-cost tasks ✅ Supported Claude Opus 4.8 1M ✅ ✅ ⭐⭐⭐ Ultra-low cost, high throughput ✅ Supported Claude Opus 4.7 1M ✅ ✅ ⭐⭐⭐⭐⭐ Best-in-class reasoning, code, long documents ✅ Supported Claude Sonnet 4.6 1M ✅ ✅ ⭐⭐⭐⭐⭐ Balanced speed and intelligence ✅ Supported Claude Haiku 4.5 200K ✅ ✅ ⭐⭐⭐⭐ Fastest response; lightweight tasks ✅ Supported Gemini 3.1 Pro 1M ✅ ✅ ⭐⭐⭐⭐⭐ Complex tasks, multimodal ✅ Supported DeepSeek-V4 Pro 64K ❌ ✅ ⭐⭐⭐⭐ Chinese comprehension, code generation Coming Soon DeepSeek-V4 Flash 64K ❌ ✅ ⭐⭐⭐⭐ Cost-effective reasoning Coming Soon Qwen3-235B 128K ✅ ✅ ⭐⭐⭐⭐ Chinese scenarios, enterprise applications Coming Soon
Model Key Features Access Status DALL·E 3 From OpenAI; rich style variety; supports non-English prompts Coming Soon Flux High-quality photorealistic style with excellent detail rendering Coming Soon Stable Diffusion Open-source; diverse styles; deployable locally Coming Soon Midjourney Outstanding for artistic creation and design styles Coming Soon
Model Type Key Features Access Status Whisper Speech-to-Text (STT) Supports multilingual audio-to-text transcription Coming Soon TTS-1 Text-to-Speech (TTS) Fast; suitable for real-time scenarios Coming Soon TTS-1-HD Text-to-Speech (TTS) Higher audio quality; suitable for content production Coming Soon
Model Type Use Case Access Status text-embedding-3-small Embedding Lightweight vectorization; suitable for large-scale text batches Coming Soon text-embedding-3-large Embedding High-precision vectorization; suitable for knowledge base retrieval Coming Soon Rerank series Rerank Re-ranks retrieval results to improve RAG accuracy Coming Soon
Use Case Recommended Models Access Status Everyday Q&A, chatbots GPT-5.5, Gemini 3.5 Flash ✅ Supported Code generation & debugging Claude Opus 4.8, Claude Sonnet 4.6 ✅ Supported Complex reasoning, math GPT-5.5, Claude Opus 4.7 ✅ Supported Image understanding GPT-5.5, Claude Sonnet 4.6 ✅ Supported Long-document processing Gemini 3.1 Pro (1M context), Claude Opus 4.7 ✅ Supported Chinese language scenarios DeepSeek-V4 Pro, Qwen3-235B Coming Soon Image generation DALL·E 3, Flux, Midjourney Coming Soon Speech recognition Whisper Coming Soon Text-to-speech TTS-1, TTS-1-HD Coming Soon
