It is getting questions like "David has 18 apples and Ivan has 7 apples. How many apples do they have together?" wrong half the time, while Gemma3 12B could very consistently answer that. Other smoke tests (like Chinese translation, and the infamous "Rs in Strawberry" test) also show poor results.

I don't know if it is a quantization/release issue, if the parameters needed for accurate responses have changed (i.e. it needs "thinking" tokens to handle its base error rate), or if the model has been so focused on audio/video that the text processing is bad.

1. A bunch of people with new Claude Code codebases in December now are working with a larger codebase, causing more context. Claude reads a lot of code files, and doesn't effectively prune from the context as far as I can tell. I find myself having to hint Claude regularly about what files to read (and not read) to avoid having 75k of unrelated files in the context window.

2. Claude Code tries to do more now, for the benefit of people who don't know exactly what they want. The trade-off is that it's worse at doing exactly what people want, when they do know. The "small fix" becomes a large endeavor for Claude.

Comments URL: https://news.ycombinator.com/item?id=47717686

Points: 1

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47717467

Points: 1

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47710907

Points: 220

# Comments: 218

Comments URL: https://news.ycombinator.com/item?id=47588549

Points: 2

# Comments: 0

The project was also already dead; the English Wikinews has had 10 "articles" posted in the last 3 weeks, two of which were trivial sports stories (a second-division Queensland football match, and the retirement of a pitcher whose last substantial year in MLB was 2019). The most recent story is that an amateur jazz group recently played at a library.

It will no longer be an attractive nuisance to the few who stumble across it. Rest in peace.

GPT-5.4-nano is less impressive. I would stick to gpt-5.4-mini where precise data is a requirement. But it is fast, and probably cheaper and better quality than an 8-20B parameter local model would be.

( https://encyclopedia.foundation/benchmarks/dashboard/ for details - the data is moderately blurry - some outlier (15s) calls are included, a few benchmark questions are ambiguous, and some prices shown are very rough estimates ).

For many "simple" LLM tasks, GPT-5-mini was sufficient 99% of the time. Hopefully these models will do even more and closer to 100% accuracy.

The prices are up 2-4x compared to GPT-5-mini and nano. Were those models just loss leaders, or are these substantially larger/better?

Comments URL: https://news.ycombinator.com/item?id=47291068

Points: 7

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47126118

Points: 1

# Comments: 0

I have tried a few systems (eSpeak, Piper, QWEN) and none of them have given satisfactory results. Huggingface seems to have no text-to-speech models with particular acclaim, either. I have been using OpenAI's gpt-4o-mini model, but that seems to be approaching end-of-life.

Is there an LLM (or non-LLM) system that you would recommend?

Comments URL: https://news.ycombinator.com/item?id=47051425

Points: 2

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47048110

Points: 2

# Comments: 1

Comments URL: https://news.ycombinator.com/item?id=47037745

Points: 1

# Comments: 0