Meta Releases Muse Spark 1.3, Its Strongest Model for Agentic and Coding Tasks
shengjia_zhao · x · 2026-09-03
Meta released Muse Spark 1.3, its strongest Spark-line model for agentic and coding tasks, available today in Muse Code and the Meta Model API.
Key points:
- Long-horizon agentic work: juggles multiple workflows in a single long thread; for open-ended goals it uses tools to build context from messy, conflicting sources, proactively corrects plan gaps, and tracks what it has learned toward a final deliverable. Trained across diverse harnesses to generalize across agentic environments.
- More active collaboration: asks clarifying questions on ambiguous prompts, calls for help when stuck, and confirms before consequential actions; adapts to preferences for frequent updates or silent background work.
- Complex instruction following: preserves detailed requirements across multi-step tasks without dropping constraints.
- Prior reasoning modes are live; max reasoning arrives after additional safety testing. Per Alexandr Wang, versus 1.2 it wastes fewer turns, uses 20% fewer tool calls and 25% fewer tokens.
Related event: Meta Launches Muse Spark 1.3: Near-Frontier Performance at ~1/8 the Price(53 posts)→
More from Models
- Open-weight TYGER speech decision model makes 80ms turn-taking calls for full-duplex voice agents — solyarisoftware · 2026-10-18
- Paying Google AI subscriber blocked: custom-domain Workspace accounts can't access AI Studio — ChrisRackauckas · 2026-10-18
- Running Ling 3.1 Flash (560B MoE, 25B Active) Across 4 DGX Sparks via llama.cpp RPC — BerryBlush_Rogers691 · 2026-10-18
- Users press OpenAI on long-delayed Adult Mode a year after it was promised — ___Patrice___ · 2026-10-18
- Polymarket opens betting market on who releases the next Gemini Argon model — Polymarket · 2026-10-18
- Grok matter-of-factly lists which groups user was racist against, exempting Africans and Latin Americans — teortaxesTex · 2026-10-18