🚨The new cloaked "Polaris Alpha" model on @OpenRouterAI appears to be an OpenAI model. It also has configurable reasoning effort (Low/Med/High). GPT-5.1 or Codex Mini are contenders for what this might be. h/t @DavidSZD1

Nov 6, 2025 · 7:58 PM UTC

7
3
4
138
Was letting me select a reasoning effort briefly but that's gone now. Seems like the "Horizon Alpha" situation we had w/ GPT-5. GPT-5.1-Instant.
1
12
Replying to @synthwavedd @OpenRouterAI @DavidSZD1
well... is it good?
1
1
seems pretty good thus far, especially for a non-reasoner creatively better than gpt-5
1
15
Replying to @synthwavedd @OpenRouterAI @DavidSZD1
Too fast for gpt-5. Many times faster than gpt-5
Replying to @synthwavedd @OpenRouterAI @DavidSZD1
营销大师 sam 总想抢 google 的风头，不过这次我看难。
1
Replying to @synthwavedd @OpenRouterAI @DavidSZD1
I don’t think it’s even reasoning when I tried
Replying to @synthwavedd @OpenRouterAI @DavidSZD1
It's GPT-5.1 but nothing crazy