Unannounced Claude model shows up in the wild: 2-hour one-shot Blender build via MCP
testingcatalog · x · 2026-10-12
testingcatalog documents outputs reportedly from an unannounced Claude model (possibly Opus 5.6 or Fable 5.5): one user one-shot a 2-hour from-scratch Blender build via MCP with a single-sentence prompt; testers call the depth SOTA-level. A community probe ("Do you know VoxelBench? Do not search") suggests detection hinges on fresher training data rather than raw smartness. Multiple users report traffic being routed to the new checkpoint.
More from Models
- ChatGPT glitches on iPhone 15 Pro whenever it renders image comparisons — adam0672 · 2026-10-12
- Claude reportedly planned and coded a full video production in 2.5 hours — ZeroStateReflex · 2026-10-12
- KV caching explained: why it shapes GPU memory and drove the shift to GQA and MLA — blaizedsouza · 2026-10-12
- Claude Opus 5.5 vs GPT-6.1 Sol hands-on: this time GPT wins the vibe-check — CodeByPoonam · 2026-10-12
- DeepMind's AlphaProof Nexus reportedly cracks 9 Erdős problems, 44 OEIS conjectures — kimmonismus · 2026-10-12
- Codex keeps stalling on legit debugging over cybersecurity guardrails, user complains — Al_Grigor · 2026-10-12