text-to-text
GPT-OSS-120B
OpenAI's flagship open-weight model has strong agentic capabilities, but is also robust in search, summarization, data analysis, support, and more.
All Available on GroqCloud
Unlock State-of-the-Art Models
Intelligent models that won't break your business.
- Support for LLMs, STT, TTS, and image-to-text models
- Industry standard frameworks and integrations
Trending Models on Groq
text-to-text
Kimi K2 0905
Moonshot's latest Kimi is known for code generation and agentic capabilities, but is also strong in creative storytelling.
text-to-text
Compound
Groq-only model brings the power of several models with tool orchestration, all through a unified API call.
speech-to-text
Whisper Large v3
OpenAI's most capable speech recognition model, delivering state-of-the-art accuracy across a wide range of audio conditions and languages, from voice agents to transcription.
text-to-speech
PlayAI-TTS
This model generates human-like audio to support writers, game developers, and content creators with interactive storytelling, narrative development, and agentic experiences.
text/image-text
Llama 4 Maverick
Meta's natively multimodal model with vision understanding, great for applications like chat, image recognition, and coding.
everything you need
15+ Integrations
Groq doesn't just serve models. It delivers end-to-end support across the AI stack, from tool calling and web search, prompt caching, batch workflows, and more.
Vision Capabilities
Fast multimodal models for image processing for tasks like visual Q&A, captioning, and OCR.
Reasoning Models
Control the reasoning presentation, including reasoning format and effort.
Agentic Tooling
Built agentic systems with a growing suite of tools, from code execution to browser automation.
OpenAI Compatible
Yep. Meaning it’s easy to configure your existing applications and try Groq speed.
Prompt Caching (NEW)
No code changes required. Caching works automatically on requests at half the cost.
Responses API
OpenAI-compatible. Integrate advanced conversational AI, including multi-modal support, stateful conversations, and function calling.
Web Search
Access real-time content from the web, including citations and list of reference sources.
Code Execution
Access real-time content from the web, including citations and list of reference sources.
Browser Automation
Control up to 10 browsers simultaneously to power parallel web research and analysis.
Batch Processing
Process large-scale workloads asynchronously with our Batch API at 50% of the cost, 24/7.
LoRAs
Upload your existing LoRA adapters to run specialized inference, all on Groq’s infrastructure.
Remote MCP
The universal bridge for connecting to external systems like databases, APIs, and tools.
Model Migration
Learn how to easily migrate prompts on commercial models to open-source ones.
Spend Limits (NEW)
Control your API costs with automated spend limits and proactive usage alerts.
25+ Integrations
Enhance your applications on Groq by integration with CrewAI, LangChain, Exa, Browserbase, and more.
Frequently Asked Questions
Migrating to Groq is designed to be seamless. You can either use one of our client SDKs, or, if you're coming from OpenAI, integrate Groq directly into your codebase.
Groq supports a growing array of popular open-source large language models - see the full list here. We are continuously expanding our model offerings to include the latest releases.
Yes! We believe you need to experience Groq's performance firsthand, so we offer a generous free tier that includes access to all of our models.
You shouldn't have to compromise on performance or budget. We offer transparent, industry-leading pricing designed to scale efficiently with your usage, ensuring exceptional value for the speed and quality you receive.