Ok but where is the product? Aren't most people just using ChatGPT from a chat prompt like they can do with any other LLM?

For whatever reason, qwq on a Macbook under ollama & Open Web UI seems to often "stall out" for me. It just never reports it even started to answer. Have you experienced this at all?

I don't have this issue with qwen2.5-coder, deepseek, etc.

Are you chatting with it directly or using it with a coding tool or IDE?

DeepSeek V3 is easily the best cost / performance non-reasoning model on the market but accessing it via API is not straightforward -- at least for people who want their model hosted in the US. For the layperson OpenAI / Anthropic are much more accessible.

Some of OpenRouter's [0][1] providers seem bunk; their version is corrupted. Occasionally just get mangled outputs. Fireworks [2] is my best experience so far: their serverless API, haven't messed with custom deployments. Fast, performant, better than other models in its size category and less expensive.

[0] https://openrouter.ai/deepseek/deepseek-chat:free

[1] https://openrouter.ai/deepseek/deepseek-chat

[2] https://fireworks.ai/models/fireworks/deepseek-v3

There's a lot of money riding on their being a lot of headroom left on LLM performance and usefulness. I personally have my doubts you can get their with the Chinese Room model of LLMs but that's where all the money and hype is riding.