Testing Claude Opus 5.5 👀
This one is interesting. the model is clearly capable, but unfortunately it didn't nail the whole thing with a one-shot prompt. had to make a few adjustments before getting it to this point. personal take: 7/10 still, DeepSeek 4.1 Flash did better
I’ve STOPPED USING GPT-Astra and Opus 5.5 as of right now
Over the last few days, i’ve been actively using GPT-Astra and Opus 5.5, but the usage limits only lasted 4–6 hours i’m totally fed up with this, so i’ve decided to switch to completely FREE models: Deepseek v4.1 flash,
Alright, so I just got OrcaSAQ-2-27B (Qwen3.8-27B quantized by orcarouter) up and running on an Nvidia P100.
Obviously, performance isn't the greatest, but I honestly didn't expect to run a modern-architecture model using trellis-based quantization on a 2016 GPU with no Tensor
Jina released a new OCR model aimed at reading difficult pages faster and more accurately.
They started with DeepSeek OCR, trained it more on difficult scans, tables, and formulas, and collected and generated examples of the mistakes it made. They gave partial credit for
You have reached the end of the archive
All of deepseek