AI News, Research and Tool Reviews
-

Runway Gen-4: myths vs facts about commercial licensing
Runway Gen-4 offers professional creators character consistency through reference conditioning, though users must manage credit costs like the 12 credits per second required for Gen-4.5. Commercial usage rights…
-

Groq LPU Supply Chain Risks and Foundry Partnerships
Groq’s $20 billion NVIDIA deal left the company financially reset but technically depleted, forcing a $650 million raise to rebuild its inference…
-

Getting Started with Groq’s LPU Inference Engine
Groq’s LPU uses a deterministic architecture to accelerate Llama inference, achieving over 1 TeraOp/s/mm2 density. The platform offers low-latency access via GroqCloud,…
-

The people shaping Mistral AI’s pivot to enterprise
Mistral AI targets €1 billion in annual recurring revenue by 2026, driven by a hiring strategy that prioritizes infrastructure veterans over AI…
-

Crusoe Energy’s $30.9B valuation and flared-gas data center expansion
Crusoe Energy Systems raised $3.9 billion at a $30.9 billion valuation to expand its flared-gas data centers. The company has a 40-gigawatt…
-

Databricks $10B Series J and AI Revenue Growth
Databricks reached a $5.4 billion annualized revenue run-rate with $1.4 billion from AI products, maintaining over 80% gross margins. The company’s valuation…
-

Lighter’s zk-perp exchange architecture and $232B monthly volume
Lighter, a decentralized perpetual futures exchange, processes tens of thousands of orders per second using Plonky2 circuits. The platform reached nearly $232…
-

Replicate vs Baseten: Production ML Deployment
Baseten charges roughly $6.50 per hour for H100 GPUs, favoring steady-state workloads with strict CI/CD governance. Replicate uses per-second billing and Cloudflare…
-

Cerebras Systems history from wafer-scale startup to AI training challenger
Cerebras Systems evolved from a 2015 startup into a major AI hardware provider by developing wafer-scale processors. The company raised $5.55 billion…
Trending Now
- Runway Gen-4: myths vs facts about commercial licensing
- The transition to agentic workflows and multi-model architectures
- Waymo scales robotaxi operations amid economic and regulatory pressures
- What changed in Leonardo.ai’s Phoenix model architecture this month
- Accelerating AI workflows with Positron photonic hardware



