The old moats are falling. The old rent seekers are losing their sway.

You love to see it.

Consider the primary use case for Anthropic: coding. Well, if you're using Claude Code or some variant, you can easily use OpenRouter to simply use whatever free tokens you can get. You may not even care what LLM you're using under the hood for a given task.

That's simply not a trillion dollar moat.

Can Anthropic jack up the prices? No. There are alternatives. There are open weight models that you can run on cloud platforms and increasingly you can run them locally. Local LLMs are going to be incredibly accessible in 3-5 years. Even now you can run Flash models on something you can buy from Apple. The inference is a lot slower compared to cloud usage but that'll change.

I see the Anthropic IPO as an effort to cash out before it all comes crashing down.

I use Paseo and there is a noticeable difference when GPT-6.1 Sol is driving the orchestrator vs Opus/Sonnet 5.5 driving the orchestrator. Anthropic models ignore the shared system prompt to use the Paseo tools for sub-agents and prefer to use its own native sub-agents. It ignores the skills mentioned in the system prompt, but will use them if I explicitly tell it to in a user prompt or I extra emphasize the instruction in an XML fenced block in the system prompt or CLAUDE.md like <BOOTSTRAP_MUST_EXECUTE_AT_START>

Sometimes, I feel like I have to explicitly prompt Anthropic models to behave the way I want it to (even with system prompts in place with the EXACT instruction). Sometimes it even feels like it's disregarding CLAUDE.md. To me, this makes me feel like it is the less compliant of the two because I constantly feel like I have to steer it manually.

That said, I prefer Anthropic models generally only when the task is more open ended and exploratory. The Anthropic models' "mind of their own" benefits here; GPT models tend to be too agreeable; and you too eager to agree with any change in direction. But for actual coding once a direction/plan is in place, I would much rather delegate it to GPT (more likely to draw on the harness for context, skills, instructions, and tools).

Anthropic's 1m context window default and shorter cache TTL are also not great in agentic flows where an agent could become idle and then re-activated. Burns tokens very quickly. I have it tuned down to 300k across all models (recommend doing this for multi-agent coding).

4-Carbon is, per benchmarks, said to be superior to Opus 5.5 at coding.

At general tasks -- search, medical, interfacing with maps and mail, planning, etc. -- the Google models are already superior.