OpenLLM V2 is here!
- Chat with all your connected models directly from the web app
- Generate images directly in your CLI of choice and access them across devices
- Track how much you save with subscriptions compared to API pricing
- Easily compare outputs between models
Just a little demo showcasing our wild v1 surface support ( chat/media ), it may be @ChatGPT@grok@cursor_ai@claudeai whichever you got or api if you prefer that, you get the full capabilities of them all in a single aggregator @OpenLLMsh
Big day for new models!
Opus 5.5 from Claude
GPT6 Sol, Terra, Luna from OpenAI
Make sure to update your fallback chains via the web app or directly through your OpenLLM connected agent
We've started tracking what subscriptions give in API value per tier
The more people who use OpenLLM, the better tracking will get
Can you believe Codex 20x gives you over $15k worth of API credits?
Your AI subscriptions are connected on your desktop. You鈥檙e working from your laptop.
OpenLLM鈥檚 inference tunnels let you send requests between your devices without copying subscription credentials around.
Use the setup you already have.
When Codex, Claude Code, or whichever subscription provider goes down, you never feel the pain with OpenLLM
Fallback chains immediately start calling up the next LLM in line
The @deepseek_ai harness plug-in is here!
One npm install and your dsh agents run on all your subs and API keys through OpenLLM, with fallback chains, unified memory and context built in.
Full setup below
The @deepseek_ai harness plug-in is here!
One npm install and your dsh agents run on all your subs and API keys through OpenLLM, with fallback chains, unified memory and context built in.
Full setup below
We're like the uber eats of gateways, because we serve you where you are.
Love using Claude Code within your terminal in Cursor?
Maybe Hermes on your VPS?
No problem
Soon you'll even be able to use OpenLLM within your Codex or Claude desktop app!