Inference

Everything Enterprise DNA has published on Inference, 10 articles, guides, insights, and news across data and AI in business.

2 blogs 8 newss

News 18 Sept 2026

An AI Designed a Chip in Two Weeks That Beats NVIDIA

Architect Labs' AI built the Redwood inference chip from a written spec in under 2 weeks, delivering 3.4x better performance per watt than NVIDIA Jetson.
News 14 Aug 2026

OpenAI Ultrafast: GPT-5.6 Sol Hits 750 Tokens Per Second

OpenAI previews Ultrafast, a new API tier powered by Cerebras that runs GPT-5.6 Sol at 14x standard speed for real-time enterprise use cases.
News 12 Aug 2026

IBM Backs Together AI With $240M to Cut Enterprise AI Costs

IBM and Together AI announced a $240M multi-year agreement to build a large-scale open-source AI inference cluster on IBM Cloud using NVIDIA HGX B300 hardware.
News 8 Aug 2026

AMD Acquires Taalas: Chips That Bake AI Models Into Silicon

AMD bought startup Taalas, whose chips encode AI model weights into silicon for 10x+ inference speedup — AI infrastructure is maturing fast.
News 18 July 2026

Fireworks AI Raises $1.5B on Specialized Enterprise AI

Fireworks AI closed a $1.5B Series D at a $17.5B valuation, showing enterprises are moving past generic models toward custom AI built on their own data.
News 9 July 2026

SambaNova Raises $1B to Compete With Nvidia on AI Inference

SambaNova's Series F signals investors are betting big on Nvidia alternatives as JPMorgan and SoftBank back on-premises AI inference at scale.
Blog 26 June 2026

Fireworks AI: What Engineers Actually Found

A practitioner's take on Fireworks AI inference in production, covering latency wins, cost surprises, and where it fits next to Together, vLLM, and OpenAI.
Blog 25 June 2026

Groq Speed: What Engineers Actually Found

Honest look at Groq inference speed in production. Latency numbers, cost surprises, where it works, where it breaks, and what teams pair it with.
News 24 June 2026

OpenAI's Jalapeño Chip Cuts AI Inference Costs by 50%

OpenAI and Broadcom unveiled Jalapeño, a custom inference chip built in nine months that delivers 50% cost savings over standard AI GPUs.
News 3 May 2026

Anthropic Eyes UK Startup Fractile for Inference Chips

The Information reports Anthropic is exploring a fourth chip supplier with Fractile's SRAM-based design that claims 25x faster inference at one-tenth the cost.