o3-mini

Fast, reasoning model with high performance.

o1

Advanced reasoning model for complex math/coding.

o1-mini

Fast, lightweight reasoning model for coding.

o1-preview

Preview of OpenAI's advanced reasoning capabilities.

gpt-4o

Most capable model, high-speed multimodal reasoning.

gpt-4o-mini

Highly cost-effective model for high-frequency tasks.

gpt-4-turbo

Latest GPT-4 Turbo model release.

gpt-4

Classic GPT-4 model, stable and capable.

gpt-3.5-turbo

Standard GPT-3.5 Turbo for simple tasks.

gpt-3.5-turbo-instruct

Direct instruction-following completion model.

whisper-1

High-fidelity multilingual speech transcribing.

tts-1

Text-to-speech engine optimized for speed.

tts-1-hd

High-quality text-to-speech audio generator.

text-embedding-3-small

Generate low-dimension semantic embeddings.

text-embedding-3-large

Generate high-dimension semantic embeddings.

text-embedding-ada-002

Legacy standard text embedding model.

omni-moderation-latest

Fast content moderation check.

babbage-002

Base completions model for simple finetuning.

davinci-002

Base completions model for advanced finetuning.

Auto / Авто - Groq

Tries top Groq models first, then falls back if one fails.

Llama 3.3 70B (Versatile)

Our most capable model for complex reasoning, coding, and writing.

Llama 3.1 8B (Instant)

Extremely fast, lightweight model for everyday tasks.

Claude 3.7 Sonnet

Anthropic's newest model, with hybrid reasoning/thinking.

Claude 3.7 Sonnet (Thinking)

Claude 3.7 with extended thinking budget enabled.

Claude 3.5 Sonnet

Anthropic's most intelligent model, top coding and reasoning.

Claude 3.5 Haiku

Fast, cost-efficient, and highly intelligent Haiku model.

Claude 3 Opus

Classic powerful reasoning model for complex tasks.

o3-mini

OpenAI's latest high-speed reasoning model.

o1

State-of-the-art reasoning model for coding and STEM.

o1-mini

Fast, lightweight reasoning model for coding.

o1-preview

Preview of OpenAI's advanced reasoning capabilities.

gpt-4o

Flagship multimodal model, highly creative and fast.

gpt-4o-mini

Cost-efficient, high-speed lightweight chat model.

Gemini 2.5 Pro

Google's flagship model with advanced multimodal vision.

Gemini 2.5 Flash

Google's high-speed, cost-efficient multimodal model.

deepseek/deepseek-r1

Flagship reasoning model with full chain of thought.

deepseek/deepseek-r1:free

Free version of DeepSeek R1 reasoning model.

DeepSeek V4 Pro

Next-gen DeepSeek V4 model optimized for coding/math.

DeepSeek V4 Flash

Ultra-fast, lightweight version of DeepSeek V4.

DeepSeek V3.2

Finetuned version 3.2 of DeepSeek chat.

DeepSeek V3

State-of-the-art conversational LLM from DeepSeek.

Llama 3.3 70B

Meta's flagship open-weights model, highly capable.

Qwen 2.5 72B

Alibaba's top performing model, expert in math/reasoning.

Auto / Авто - Gemini

Tries top Gemini models first, then falls back if one fails.

Gemini 2.5 Pro

Google's flagship model with advanced multimodal reasoning.

Gemini 2.5 Flash

Google's high-speed, cost-efficient multimodal model.

Auto / Авто - OnlySQ

Tries top OnlySQ models first, then falls back if one fails.

gpt-5

OnlySQ's top tier simulated or next-gen GPT-5.

claude-sonnet-4-6

Next-generation Claude model with extreme intelligence.

deepseek-r1

Advanced reasoning model with deep chain of thought.

gemini-3.5-flash

Ultra-fast multimodal model.

grok-4.20

Grok's high performance model.

llama-3.3-70b

Meta's flagship open weight model.