5 days left of my Venice.ai subscription, then a new stack:
- @PayPerQ APIs (probably with NWC auto top-up)
- Open WebUI for chats with SearXNG for web search
- OpenCode for projects
I think it offers a good alternative when you can't afford new hardware.
For the models, I'm planning a mix of cheap and intelligent, depending on the task:
- gpt-oss-120b
- GLM 4.7 Flash
- MiniMax M3
- Gemini 3.1 Pro Preview
- Qwen3.7 Max
Almost all set 😎
My only apprehension is to have to pay for EVERY prompt moving forward. Until now, my prompts were basically free. I'd throw anything in the chat box. And I've never thought, "Oh, I shouldn't send this because it'll cost money."
- @PayPerQ APIs (probably with NWC auto top-up)
- Open WebUI for chats with SearXNG for web search
- OpenCode for projects
I think it offers a good alternative when you can't afford new hardware.
For the models, I'm planning a mix of cheap and intelligent, depending on the task:
- gpt-oss-120b
- GLM 4.7 Flash
- MiniMax M3
- Gemini 3.1 Pro Preview
- Qwen3.7 Max
Almost all set 😎
My only apprehension is to have to pay for EVERY prompt moving forward. Until now, my prompts were basically free. I'd throw anything in the chat box. And I've never thought, "Oh, I shouldn't send this because it'll cost money."
1
