Hi,

Maxime • • •
npub1zkse38pvfqlkcmcc7tw6zqecj7sqxe5lgj0u9ldylghmdjfppyqqtsa4du
hex
35e66af0c881be092a3edf108be1f8054a4b32d115197e4602a1c6d742e2a45fnevent
nevent1qqsrten27rygr0sf9gld7yytu8uq2jjtxtg32xt7gcp2r3khgt32ghcprpmhxue69uhhyetvv9ujuem4d36kwatvw5hx6mm9qgsptgvcnskys0mvduv09hdpqvuf0gqrv605f87zlkj05takeyssjqqumddlfKind-1 (TextNote)
↳ Reply to Event not found
6039932fd2e7831f9805c08c2962ece5b8dac9a72b3e7fa962db466408bbf86e...
Hi,
I agree Opus is the most intelligent. For which tasks are you claiming GLM 5.1 is not good enough though ?
I personaly mix a bit of all. Through Opencode, Claude code, Codex, and cli commands like claude -p or opencode run.
I find good result and balance between tokens consumption, api cost + minimal subsciption.
I recently Added GPT 5.4 nano and grok-4-1-fast-reasoning. They are fast, cheap and performant models.
Raw JSON
{
"kind": 1,
"id": "35e66af0c881be092a3edf108be1f8054a4b32d115197e4602a1c6d742e2a45f",
"pubkey": "15a1989c2c483f6c6f18f2dda1033897a003669f449fc2fda4fa2fb6c9210900",
"created_at": 1775763870,
"tags": [
[
"e",
"6039932fd2e7831f9805c08c2962ece5b8dac9a72b3e7fa962db466408bbf86e",
"",
"root"
],
[
"p",
"3f770d65d3a764a9c5cb503ae123e62ec7598ad035d836e2a810f3877a745b24"
],
[
"r",
"wss://relay.primal.net/"
],
[
"r",
"wss://premium.primal.net/"
],
[
"r",
"wss://relay.wellorder.net/"
],
[
"r",
"wss://relay.ditto.pub/"
],
[
"client",
"Primal Web"
]
],
"content": "Hi,\nI agree Opus is the most intelligent. For which tasks are you claiming GLM 5.1 is not good enough though ?\nI personaly mix a bit of all. Through Opencode, Claude code, Codex, and cli commands like `claude -p` or `opencode run`.\nI find good result and balance between tokens consumption, api cost + minimal subsciption.\n\nI recently Added GPT 5.4 nano and grok-4-1-fast-reasoning. They are fast, cheap and performant models. ",
"sig": "ce2ea608317a32b23cbdba5bdefe97f61970e0467685cef301fcfdd494e209ee5631a181b9378fbbcd7430373a5804b05d3284d7eda08a0b658753a3d24fadf7"
}