Every games studio is going to zero.
Don’t think so? Just watch this gameplay.
Do you feel it yet? The gaming landscape has shifted. It’s over for the big studios, soon everyone will be playing this.
Kimi K3 just 3 shotted this CS:GO × Portal clone for me using around 600,000 tokens.
$3.24 in API usage. The same token cost would be $10.80 with Fable 5 & $6 with GPT-5.6 Sol.
The era of free indie game development is closer than you think anon!
Jax added support for opaque types with custom tangents. One cool use is proper handling of batched geometric objects. Here's a short demo of a differentiable rasterization, i.e. min L2 w/ 1500 triangles to a photo. Interesting to move beyond tensors.
(https://t.co/96vYxPXIFK)
i continue to be impressed by OAI’s context window compaction, prob their largest technical moat atm. distilling information down to its purest form without sacrificing noticeable loss is an art. it’s what separates an average professor from a great one.
Turns out GPT-5.6 Sol is actually SoTA on ARC-AGI-3.
Just took two setting changes. You just have to allow it to reason and work over multiple context windows with the help of our canonical compaction implementation.
https://t.co/wHjaNsvIv8
@0xrsydn Luna on xhigh or ultra is great for some small context problems but any smaller I think I’d rather just 5.5 or possibly even 5.4 depending on the task. Sol has been great but a little too thorough with its tests, seems quite wasteful. And I don’t even want to start with Terra
5.6 terra is completely retarded and makes way too many complex decisions on its own without seeking approval. such a huge difference between sol and luna, something is seriously wrong with this model