Agentic Task: Install K3s in your own slicervm
Left: DeepSeek V4 Flash 0731 - 2x DGX Sparks Right: Qwen 3.8 27B (FP8) - 1x RTX 6000 Pro 40 (ish) tok/s gen vs 88 tok/s Why do they both feel "slow"? Thinking aka "variant" - they're spending a lot of tokens on reasoning.
Deepseek + Hermes es WIN
Pero pruebalo con Grok y vas a flipar
Call out Pumpfun clawd bot loading.
With the new deepseek_ai harness What a time to be a solana degen Make solana ai great again Powered by solana:8cHzQHUS2s2h8TzCmfqPKYiM4dSt4roa3n7MyRLApump
Producthunt just submitted my product today and took a look at the list, which is quite interesting.
There is Gemini flash 3.7 and Deepseek Harness
You have reached the end of the archive
All of deepseek