Meet Oh-My-Hermes
I burned 200B tokens tuning my Hermes Agent setup, and now you can get all of it with a one-line install.
It comes with 150+ skills, a long-term memory system that lasts twice as long, graph engineering, and memory-management skills. It also includes subagent
omher desktop @ beta
i've already set up separate windows to spin up and work across multiple sessions like claude, codex, omo, zcode, and dsh,and now i'm thinking about expanding the interface so base hermes bots can wield and connect them as tools almost like spawning threads.
github.com/rlaope/oh-my-h…
a new version of oh-my-hermes is out.
github.com/rlaope/oh-my-h…
the reliability and speed of the goal engineering (graph) workflow have been optimized.
ai research, agi loops, lsp, and task-execution prompting skills have all been upgraded
delivering up to ~6x better performance
We're serving DeepSeek at 90% off.
DeepSeek V4 Flash standard price: $1.19 per 1M output tokens.
DeepSeek V4.1 Flash Eco on OpenGateway: $0.12.
Input: $0.40 → $0.03.
How?
We serve it ourselves.
We don't resell someone else's API. DeepSeek runs on our own GPUs, so there's no
we also run flash sales on our
GLM-5.3-Flash 500–700 decode tok/s.
opengateway.ai/models
and as a permanent offer, we have our DeepSeek flash eco model at 90% off.
if you're looking for cheap, ultra-fast models and want updates on these flash sales,follow me or @OpenGatewayai.
