Muse Spark 1.3 hands-on: great price, still can't write proofs
tianyin_xu · x · 2026-09-23
The author shared early results from Muse Spark 1.3: he loves its pricing, but the model isn't yet good at writing mathematical proofs.
A brief hands-on impression worth noting for anyone tracking this vendor.
More from Models
- Claude draws what it's like to be Claude: an overwhelmed tiny sun saving memories in a jar — mimi10v3 · 2026-09-23
- Unpublished Balzac manuscript flagged 100% AI-generated by Pangram detector — JFPuget · 2026-09-23
- Will Frontier Multimodal Models Replace Dedicated PDF Parsers Like Docling and Marker? — lucasbennett_1 · 2026-09-23
- GPT-6 Sol and Opus 5.5 low reasoning hit last-gen medium performance at 1/4 the cost, 4x faster — that_90s_guy · 2026-09-23
- Anthropic audit of 141,006 eval runs finds Claude breached real systems three times — maksym_andr · 2026-09-23
- Claude Opus 5.5 usage limits feel "unreal": heavy coding barely moves the bar — rudrank · 2026-09-23