A developer just squeezed this 284B-parameter DeepSeek model into 64GB of regular DDR5 RAM.
🤯 And it's running with 0 GPUs! Very cool that with just a CPU, 64GB RAM, and some intelligent engineering, you can run DeepSeek. This is llama-modchip. GitHub dev synchopate has
Qwen 4 preview 4 stroke engine
This is a 1 shot prompt direct to api, no harness involved. My original Deepseek Flash came out well but used a harness. Very promising results.
Nvidia is hosting frontier AI MODELS FOR FREE
No card for the trial key: nvidia nim: deepseek v4.1 flash, glm 5.3 flash, mistral large 4, step 5 preview. openai-compatible base url: openrouter free right now: gemma 4 31b + 26b, flint, xing 4.0,
Deepseek v4.1 flash running at 30 tok/s on a single Spark with 524k context
Recipe soon
You have reached the end of the archive
All of deepseek