Issues related to model selection

[tlerbao]

As shown in the figure below: I replaced "explore" and "librarian" with "gpt-5.4-mini" because I don't have access to "grok" and "minimax."

Additionally, I replaced all instances of "gpt-5.3-codex" with "gpt-5.4" (since the official announcement stated that 5.4 already encompasses and even surpasses the programming capabilities of 5.3-codex).

My question is, will these changes affect the final outcome? Do you have any suggestions?

[TomLucidor]

Please DYOR and try other models as well. Check Artificial Analysis to see if they are better (relative to cost)

[tlerbao]

If cost is not a consideration, which ones are the best? Haha

[TomLucidor]

On the same boat #1768 #1400

[tlerbao]

This project is actually being maintained by you manually checking for duplicates, not by a bot. Thank you for your hard work.

[TomLucidor]

The bot is for settling arguments, and I try to find something to solve this cus even the bots/maintainers isn't clear if this works. Everything is trial-and-error.

[ukrospm]

@tlerbao hey, can you please paste here the oh-my-opencode.json that you have, because for some reason it is not switching models for me :-/ i have 2 local models, one fast one slow and everytime it is using smart slow model.

[tlerbao]

下面是我的最新版

{
  "$schema": "https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/dev/assets/oh-my-opencode.schema.json",
  "disabled_hooks": ["session-notification"],
  "tmux": {
    "enabled": true,
    "layout": "main-vertical",
    "main_pane_size": 60
  },
  "agents": {
    "sisyphus": {
      "model": "github-copilot/claude-opus-4.6",
      "variant": "max"
    },
    "hephaestus": {
      "model": "openai/gpt-5.4",
      "variant": "medium"
    },
    "oracle": {
      "model": "openai/gpt-5.4",
      "variant": "high"
    },
    "multimodal-looker": {
      "model": "openai/gpt-5.4",
      "variant": "medium"
    },
    "prometheus": {
      "model": "github-copilot/claude-opus-4.6",
      "variant": "max"
    },
    "metis": {
      "model": "github-copilot/claude-opus-4.6",
      "variant": "max"
    },
    "momus": {
      "model": "openai/gpt-5.4",
      "variant": "xhigh"
    },
    "atlas": {
      "model": "github-copilot/claude-sonnet-4.6"
    },
    "sisyphus-junior": {
      "model": "github-copilot/claude-sonnet-4.6"
    }
  },
  "categories": {
    "visual-engineering": {
      "model": "github-copilot/gemini-3.1-pro-preview",
      "variant": "high"
    },
    "ultrabrain": {
      "model": "openai/gpt-5.4",
      "variant": "xhigh"
    },
    "deep": {
      "model": "openai/gpt-5.4",
      "variant": "medium"
    },
    "artistry": {
      "model": "github-copilot/gemini-3.1-pro-preview",
      "variant": "high"
    },
    "quick": {
      "model": "openai/gpt-5.3-codex-spark"
    }
  }
}

[ukrospm]

thank you, found my mistake <3

[nigiwen]

How was your table generated?

[tlerbao]

ai

[chen-lightspeed]

do we have to maintain our own lists or is there a way to easily pull the latest model selection? I find this hectic if we have to do this regularly