
If you start using Nori for cloud agents, make sure you use Subconscious for inference 🤝
✓The inference provider for long-running agents

If you start using Nori for cloud agents, make sure you use Subconscious for inference 🤝

Build your exploit factory and protect your product!


Sharing the good word of Subconscious at @getpostman last night. Great questions and energy from the crowd 🤝


GLM 5.2 live on the Subconscious API - Frontier coding ability - Powered by our inference system optimized for agents 95%+ cache hit rate, fast token throughput, longer context reasoning with automated nearly-lossless context compression. Check it out.

We designed an inference runtime specifically tuned to run agent workloads with a prefix *and* suffix cache, aka when you prune stuff in them middle you still get a 100% cache hit on the tokens around it. We see a 95% cache hit rate for our team's coding agent instance w/ GLM.

GLM-5.2, the latest frontier coding model, can now control and compress its own context in inference time, powered by Subconscious Cache. Model-driven context engineering will be the future of AI Agents. subconscious.dev/blog/subcon…

Won the Beat The Clock Agent Hack at Wayfair HQ. 2 hours to build. Live demo. 200+ hackers. Thanks @Wayfair, @subsysdev, @baseten, and @Cloudflare for hosting 🚀 #BosTechWeek

Last week at NY Tech Week, founders and builders filled the room for our BYOA (Bring Your Own Agents) workshop. Everyone brought new ideas and their own agents to showcase and try new tools from us and our partners. Big thank you to our co-host @Mercury, and to @paysponge and @subsysdev for joining us!

This company puts GPUs on top of water heaters, and uses the GPU to heat the water. It's an incredible company, and we're excited to power agents on this distributed inference network. More info below!

Launch of our TIM-Qwen3.6-27B Model today. We took a great, small Qwen model and substantially improved it's performance: - 1M+ context window (10x extension) - Run 3x as many concurrent runs - 75% less memory required - 1.5x token throughput - OpenAI / Anthropic SDK compatible

If you haven't tried the Subconscious playground, here's a look. I gave the model search tools, asked it to tell me about our research paper, and boom, we have a personalized deep research agent. Took 15 seconds to make.

Beyond Context Limits Subconscious Threads for Long-Horizon Reasoning

Today we're launching Subconscious: a new platform for building agents with long-horizon reasoning and tool use, backed by MIT research. One API call. Tool use. Context beyond existing limits. If you're building agents, let's talk.

Hello World