Testing DeepSeek, OMP and OpenCode harnesses using the same Ox Alpha model.
I gave all 3 the same Ranger RF-31D reference image from Interstellar and the exact same one-shot prompt to see how well each harness could recreate it. Here’s the result 👇
There is something they often notice about NVIDIA
You can try out a large number of AI models via a free API from It includes examples of: Qwen Llama Gemma DeepSeek And others Choose the model Generate API Key And use it directly The interface is compatible with the OpenAI API This means that you can often install it easily in the existing tools…
Just imagine, you go and put this tool on the table.
You will sit in front of it and not touch the machine for 5 hours. Then you will open it and install it gently and politely. You will run DeepSeek-R1 671B Q4 with 128K context and make a lot of parallel inference.
A couple of days ago we ran Qwen3.8 27B on a laptop's Qualcomm NPU.
It was never meant to run there. It was not built for that silicon, nobody quantized it with an NPU in mind, and it ran anyway. Here are some of the other models we run on it: deepseek_ai's DeepSeek R1.
You have reached the end of the archive
All of deepseek