Groq
llama-3.3-70b-versatile
OpenAI
GPT-4o
Flagship · Multimodal · Fastest reasoning
4o
GPT-4o Mini
Cost-efficient · Fast · Capable
Mini
o1-mini
Chain-of-thought reasoning
o1
DeepSeek
DeepSeek Chat (V3)
Strong reasoning, coding & analysis
V3
DeepSeek R1
Chain-of-thought · Deep reasoning
R1
Hugging Face — Open Models
Llama 3.2 3B Instruct
Meta · Fast, free inference
FREE
Llama 3.1 8B Instruct
Meta · Balanced power & speed
FREE
Mistral 7B Instruct v0.3
MistralAI · Efficient & capable
FREE
Qwen 2.5 7B Instruct
Alibaba · Multilingual, strong code
FREE
Custom HF Model
Enter any model ID from huggingface.co
CUSTOM
Groq — Lightning Fast FREE TIER
Llama 3.3 70B
Meta · 300+ tok/s · Best free model
FAST
Llama 3.1 8B Instant
Meta · Ultra-fast responses
⚡
Mixtral 8x7B
Mistral AI · MoE · 32k context
32K
Gemma 2 9B
Google · Efficient & capable
FREE
OpenRouter — Multi-Provider FREE
Llama 3.3 70B (Free)
Meta via OpenRouter · No cost
FREE
DeepSeek R1 (Free)
Chain-of-thought reasoning · No cost
R1
Gemini 2.0 Flash (Free)
Google · Fast & multimodal
FREE
Loading
SAI
Welcome Back, User
Ready to chat — powered by Groq · Llama 3.3 70B by default. Add your API keys in Settings to unlock more models.
No API key configured. to start chatting.