Alec

OpenClaw just got expensive. Here's what actually changed and what to do about it. Anthropic blocked Claude subscription from being used with OpenClaw. That means you can no longer pay a flat $200/month and get unlimited Opus access for your agents. For anyone who doesn't know: OpenClaw turns AI into a digital worker. It codes, replies, schedules, builds, all running in the background while you do other things. Opus was the engine behind it. Best reasoning, best planning, best at running other models. Now you pay per use through the API. No more flat rate. So smart builders are splitting the workload: 1. Opus API for decisions and planning only (the expensive part, used sparingly) 2. ChatGPT for coding tasks (most people already have this) 3. Free/cheap models like Gemma, Qwen for background tasks, scans, scheduling, simple stuff You don't use the most expensive model for every small task. You route each task to the right model for the job. My take: This is going to separate two groups very quickly. People who understand how to build with AI agents properly. And people who just used it because one subscription made it easy. The flat fee made it simple. Now it takes real understanding. If you're serious about using AI to run parts of your work, this is the setup to learn. Not optional anymore.

Anthropic just banned Claude subscriptions from powering OpenClaw. Here's why my stack was already built for this. I never ran Opus 4.6 through a subscription for OpenClaw or Hermes. It runs in Claude Code for complex external dev only. Same with GPT-5.4 in Codex. The internal agent runtime is a completely different stack: 1. Qwen3.5 9B runs locally. $0. Always on. Feeds the subconscious ideation loop 24/7. Beats GPT-OSS-120B by 13x. Awesome. 2. MiniMax M2.7 is the agent's backbone. 97% skill adherence, built for agents, $0.30/M tokens. The $10 plan allows for 1500 calls every 5 hours. Amazing. 3. GPT-5.4 mini is the Hermes brain. debates ideas with the subconscious, builds output, ~$0.075 avg per run. It's smart enough to orchestrate your entire system, and you can actually use your subscription plan here via OAuth. Incredible! Over the last 24 hours, the subconscious ran 15 times, for a total of $1.58. Not too shabby for an always-improving agentic system. The lesson is to build your agent stack on a multiple LLM stack. Local models handle volume. Generous subscription models handle execution and judgment. You own the cost structure. Full-stack breakdown in the table. (see image)

So… What are people using for their open claw now? Or are you just paying for the usage? And if so, how much we racking up in costs there