Quote · Nerd Snipe with Theo and Ben
5 different models dropped last week & the GPT-5.6 usage limits are brutal
Where this was said
Muse Spark 1.1: The Least Exciting Drop
At 16:00 · chapter starts 12:55
Kicking off the model rankings from least to most exciting, Theo and Ben land on Muse Spark 1.1 as the easy last place. It's only available through the API, benches around Opus tier, and yet carries the eerie feel of earlier open-weight models — competent on well-defined, familiar tasks, but liable to loop and hallucinate on anything vague or exploratory. Ben likens it to the 'Qwen behavior' of reasoning in circles. The most cutting observation: Meta has been training their models on employee screen recordings, which means this model is, as Theo jokes, state-of-the-art at one thing — applying for jobs at Anthropic. The bigger structural point is sobering: Meta's ML engineers apparently spend over half their time labeling data, turning their engineering talent into a data pipeline rather than a research team.
Grok 4.5 is the first model outside OpenAI and Anthropic that can handle multi-part instructions without getting lost or confused. At $2 in / $6 out and 102 TPS on OpenRouter, it's not frontier-level intelligence but it's genuinely usable — and for the first time, that's enough.
Grok 4.5 was averaging 102 tokens per second on OpenRouter at the time of recording, indicating strong throughput for an affordable non-frontier model.
Grok 4.5 is priced at $2 per million input tokens and $6 per million output tokens on OpenRouter, making it one of the cheapest models capable of serious developer tasks.
