Ollama has become the largest developer platform for open models, with 8.9m developers & one million new installs each week. Today we announced our investment in the Series B.
An agent that forgets mid-sentence isn't a context-size problem. It's a memory-architecture problem. A preflight step, a skills library, & a watchdog fix it.
Forward-deployed engineering has gone from a Palantir signature to an industry default. The total bill : nearly $10b in 12 months.
The Economist put 25 frontier AI models through the World Values Survey. The variance is real, the obvious groupings don't hold, & worldview may be the next line item on an AI RFP.
Prioritize routing over model choice. Most AI work runs on cheap local models.
Anthropic spends 2.3x payroll on compute. Top software firms spend 0.4x. Three scenarios for where the rest of the market lands by 2029.
The inference market is the largest in software. We invested in Sail's Series A to build queuing infrastructure for long-horizon agents that run for hours.
Reselling inference is a zero-margin commodity. The margin question comes down to cost-plus versus value-based pricing : one anchors to the inference line, the other decouples from it.
Databricks runs at $6.9b ARR growing 80%, widening the gap over Snowflake to $1.6b. Its AI products sit on the token path, where companies grow explosively even at scale.
A 500+ comment HN discussion reveals the current state of local AI coding - Qwen 3.6 35B dominates model mentions while Pi leads agents.
