Celeris-1 Magnus: New Model Claims Top Spot on τ³-bench for Agentic Work
timshi_ai · x · 2026-09-01
Celeris introduced Celeris-1 Magnus, a hybrid diffusion model built specifically for agentic workflows derived from Qwen3.8-27b. On the τ³-bench banking suite (97 tasks), Magnus achieved a 41.2% solve rate with a 55-second median time, outperforming GPT-5.6-sol (38.1%) in both speed and accuracy. It features a "reasoning dial" that boosts performance by 13.4 points when enabled. The API is OpenAI-compatible, allowing seamless integration for existing agents.
More from coding & agent
- LLM one-shots a Minecraft 3x3 piston door by building its own Linux test infra — Vjeux · 2026-10-18
- Dev builds Mumbai-inspired 3D manga background generator from OpenStreetMap data — CurieuxExplorer · 2026-10-18
- Solo founder builds a money-printing data service on Claude's Pi Durable harness — MicahBerkley · 2026-10-18
- Dev builds Arch-based OS with custom desktop and browser using Claude Code — Snoo-10493 · 2026-10-18
- Five SaaS agent workflows fail the same way — and it's never model intelligence — usehive · 2026-10-18
- Claude skill generates hand-drawn interactive diagrams that teach kids how things work — CatAstro_Piyush · 2026-10-18