DeepSeek v4.1 Flash at 378 tok/s 99.7% Cache hit rate r runinfra.ai 2 OsamaJaber 1 month ago 0 comments
Agent loops stop re-reading your codebase on every turn r runinfra.ai 2 OsamaJaber 2 months ago 0 comments
DeepSeek V4 Pro at 207 tok/s with the full 1M context, no quantization r runinfra.ai 3 OsamaJaber 2 months ago 0 comments