Llama 3 Now Available to Developers via GroqChat and GroqCloud™ Here’s what’s happened in the last 36 hours: April 18th, Noon: Meta releases versions of its latest Large Language Model (LLM), Llama 3. April 19th, Midnig…
AI generation is temporarily unavailable; this entry was preserved with deterministic fallback metadata.
Llama 3 Now Available to Developers via GroqChat and GroqCloud™ Here’s what’s happened in the last 36 hours: April 18th, Noon: Meta releases versions of its latest Large Language…
Jonathan Ross, CEO and Founder of Groq, and Tareq Amin, CEO of Aramco Digital, a subsidiary of Saudi Aramco, proudly announced their partnership to build the largest AI inference data center in Saudi Arabia. The data ce…
AI generation is temporarily unavailable; this entry was preserved with deterministic fallback metadata.
Jonathan Ross, CEO and Founder of Groq, and Tareq Amin, CEO of Aramco Digital, a subsidiary of Saudi Aramco, proudly announced their partnership to build the largest AI inference…
What is Context Length? As businesses look to leverage Large Language Models (LLMs) for conversational AI, generative AI, and analytics, a crucial factor often gets overlooked: context length, also known as context wind…
AI generation is temporarily unavailable; this entry was preserved with deterministic fallback metadata.
What is Context Length? As businesses look to leverage Large Language Models (LLMs) for conversational AI, generative AI, and analytics, a crucial factor often gets overlooked: co…
This blog is adapted from an original post by Groq CEO and Founder, Jonathan Ross. If Large Language Models (LLMs) are the printing press of the Generative AI age, then what’s next? What will be the AI equivalent of the…
AI generation is temporarily unavailable; this entry was preserved with deterministic fallback metadata.
This blog is adapted from an original post by Groq CEO and Founder, Jonathan Ross. If Large Language Models (LLMs) are the printing press of the Generative AI age, then what’s nex…
The rapidly evolving field of Artificial Intelligence (AI) has led to significant advancements in Machine Learning (ML), with "inference" emerging as a crucial concept. But what exactly is inference, and how does it wor…
AI generation is temporarily unavailable; this entry was preserved with deterministic fallback metadata.
The rapidly evolving field of Artificial Intelligence (AI) has led to significant advancements in Machine Learning (ML), with "inference" emerging as a crucial concept. But what e…
Today, one year this week since we launched GroqCloud, we're celebrating the one million developers who are now on GroqCloud! In just a year this community of builders, makers, and innovators have shipped incredible app…
AI generation is temporarily unavailable; this entry was preserved with deterministic fallback metadata.
Today, one year this week since we launched GroqCloud, we're celebrating the one million developers who are now on GroqCloud! In just a year this community of builders, makers, an…
Overview Groq LPU™ AI Inference Technology Groq builds fast AI inference. Groq® LPU™ AI inference technology delivers exceptional AI compute speed, quality, and affordability at scale. Groq AI inference infrastructure,…
AI generation is temporarily unavailable; this entry was preserved with deterministic fallback metadata.
Overview Groq LPU™ AI Inference Technology Groq builds fast AI inference. Groq® LPU™ AI inference technology delivers exceptional AI compute speed, quality, and affordability at s…
GroqCloud™ provides fast inference for complex AI solutions that require instant responsiveness. But what happens when your use cases expand and require features beyond speed? That’s where Batch Processing comes in –now…
AI generation is temporarily unavailable; this entry was preserved with deterministic fallback metadata.
GroqCloud™ provides fast inference for complex AI solutions that require instant responsiveness. But what happens when your use cases expand and require features beyond speed? Tha…
You know Groq runs small models. But did you know we run large models including MoE uniquely well? Here’s why. The Evolution of Advanced Openly-Available LLMs There’s no argument that Artificial intelligence (AI) has ex…
AI generation is temporarily unavailable; this entry was preserved with deterministic fallback metadata.
You know Groq runs small models. But did you know we run large models including MoE uniquely well? Here’s why. The Evolution of Advanced Openly-Available LLMs There’s no argument…
Appointment supports Groq’s global expansion and growing AI inference demand Mountain View, Calif. — September 22, 2025 — Groq, the global pioneer in AI inference, today announced the appointment of Simon Edwards as Chi…
AI generation is temporarily unavailable; this entry was preserved with deterministic fallback metadata.
Appointment supports Groq’s global expansion and growing AI inference demand Mountain View, Calif. — September 22, 2025 — Groq, the global pioneer in AI inference, today announced…
GroqCloud announces beta availability of remote Model Context Protocol (MCP) server integration, enabling faster, lower-cost AI applications with seamless tool connectivity and zero-code migration from OpenAI.
Remote MCP integration allows AI models to interact with external tools via OpenAI-compatible API.
Compatible with OpenAI Responses API and remote MCP spec, requiring no code changes.
Groq announces two key updates for its GPT-OSS models: price reductions and prompt caching, aimed at improving cost efficiency and speed for AI inference. New pricing is effective immediately and retroactive to October 2025 invoices. Prompt caching offers up to 50% discount on cached tokens, lower latency, and higher rate limits with zero configuration.
Price reductions for GPT-OSS models, effective immediately and retroactive to October 2025.
Prompt caching launched, offering 50% discount on cached tokens and reduced latency.
Based on practical experience, this guide explains how to reliably integrate open-source LLMs into products. The core is a four-step loop: Read (only necessary context), Constrain (clear system and formatting rules), Act (structured outputs, function calls, or plain text), Explain (show users steps and citations). It covers common patterns (router, extractor, translator, etc.), safe shipping (testing, monitoring, fallbacks), and common pitfalls. The goal is to build invisible, reliable AI features that users depend on daily.
The best AI features are often invisible, letting users complete tasks without noticing AI.
The core workflow is a four-step loop: Read, Constrain, Act, Explain.
GroqCloud announces day zero support for OpenAI's GPT-OSS-Safeguard-20B, a new open-source safety-classification model running at over 1000 t/s. Key features include bring your own policy, configurable reasoning effort, full reasoning trace, prompt caching, and 128k token context window. Pricing matches the base GPT-OSS-20B model.
OpenAI releases GPT-OSS-Safeguard-20B, fine-tuned from GPT-OSS-20B for safety classification.
GroqCloud provides day zero access with inference speeds over 1000 t/s.
Groq announces MCP Connectors in beta on GroqCloud, starting with Google Workspace. These pre-built, Groq-hosted MCP servers enable AI agents to interact with Gmail, Drive, and Calendar via the Responses API without managing your own MCP server.
GroqCloud launches MCP Connectors beta, initially supporting Google Workspace.
Drop-in compatibility, zero deployment burden, low latency, and low cost.
Groq has been recognized as a Cool Vendor in the 2025 Gartner AI Infrastructure report, highlighting its LPU chip's deterministic, low-latency inference that scales linearly. Over 2.5 million developers use Groq for up to 5x faster and cheaper performance than GPUs.
The article discusses U.S. leadership in AI compute, especially inference, and proposes an export policy that balances market flexibility with consortium coordination to maintain strategic advantage.
The U.S. dominates AI compute, controlling 74% of high-end training capacity.
Inference compute is becoming the critical bottleneck for AI deployment at scale.
GroqCloud is expanding its AI inference infrastructure globally to meet the growing demand from real-time applications moving from experimentation to production. A new UK data center, in partnership with Equinix, brings deterministic, high-performance inference closer to European developers and enterprises. GroqCloud now has over 3.5 million developers and sustained increases in production traffic.
GroqCloud surpasses 3.5 million developers with growing production traffic.
New UK data center in partnership with Equinix expands European presence.
Groq’s LPU is purpose-built for inference, achieving ultra-low latency without sacrificing accuracy through TruePoint numerics, SRAM-based memory, static scheduling, and tensor parallelism. Kimi K2 runs at 40x performance on Groq, demonstrating the architecture’s efficiency.
LPU eliminates the accuracy-speed tradeoff inherent in GPU inference
TruePoint numerics deliver 2-4x speedup over BF16 with no measurable accuracy loss