Reactions to GPT-6 Sol are very mixed so far.
On the one hand, the pricing is great. No doubt about that.
But there’s also quite a bit of negativity around its performance. In benchmarks and real-world coding, it seems nowhere close to Astra or Opus 5.5, despite OpenAI’s claims.
Is it just a souped-up Terra model after all?
I finally tested GPT-6 Sol on a real work. 2 repos. 105 hidden bugs. Find and fixed what you can.
It looks like a huge degradation. The results:
- GPT-6 Astra (max): 45
- GPT-5.6 Sol (max): 43.5
- Opus 5.5 (max): 41.7
- Muse Spark 1.3 (max): 32.2
- GPT-6 Sol (max): 29.3
Until you measure the cost (API-equivalent):
- GPT-6 Astra (max): $33.04
- GPT-5.6 Sol (max): $95.25
- Opus 5.5 (max): $58.53
- Muse Spark 1.3 (max): $18.11
- GPT-6 Sol (max): $9.93
More effort levels (xhigh, high, medium, low) dropping in this thread today 🧵