DeepSeek V4 Flash doesn’t fit on NVIDIA L4 GPUs.
The memory isn’t there. GraphX made it run on eight of them anyway. If it can do that on hardware that was never meant for the job, imagine what it does on everything else.
This is even more insane, DeepSeek-v4.1-flash build me a complete game in one-shot attempt.
Not with any harness, not using any external assets, no tool calls, just single turn, one prompt in and html output. Model is DeepSeek-v4.1-Flash-1017 (deepseek-flash is the model name
DeepSeek open-sourced DeepInfer, the engine behind its API.
Its receipt: $38 of revenue per GPU-hour (off-peak). In 2025 the same math gave ~$13 on H800s. The catch: 64 + 64 Blackwell-class GPUs, and no cache store in the repo.
You have reached the end of the archive
All of deepseek