StepFun Launches Step 5 Preview: 600B/27B MoE Flagship, Open Weights Oct 15
StepFun_ai · x · 2026-09-20
StepFun introduced Step 5 Preview, a new flagship model for agentic work claiming frontier performance in software engineering and professional knowledge work, with a finance focus.
- Specs: 600B total / 27B active MoE, 1M context + vision, substantially lower task cost at comparable intelligence
- Long-horizon execution: sustains runs up to 24 hours (GPU kernel optimization, automated post-training), iterating on execution feedback
- Professional work: one agent action coordinated 950 web fetches and assembled 300,000 monthly records across 1,000 locations over 25 years; another produced a 17-sheet analytical workbook with source reconciliation and formulas
- Finance: FinStepBench (LiveSearch, CorporateValuation, DeepResearch) plus FrontierFinance across six investment use cases
- Open weights coming Oct 15; Step Plan subscription free for a limited time
Related event: StepFun Launches Step 5 Preview: 600B MoE Flagship, Open-Weights on Oct 15(17 posts)→
More from coding & agent
- Ora's Ranker 3.0 grades whether agents can actually use your site; scores drop 40+ points — yenkel · 2026-11-12
- Gemini API Managed Agents now support self-hosted sandboxes, incl. Vercel — steren · 2026-11-12
- Rayline launches ARC, a per-turn model router claiming 90% cost cuts — testingcatalog · 2026-11-12
- Karpathy's llm-wiki: a pattern for building personal knowledge bases with LLM agents — stochasticchasm · 2026-11-12
- Agent self-improvement with frozen weights: failure traces add just 4pp, blind search gets 60% — burny_tech · 2026-11-12
- Microsoft on agentic schedulers: LLM proposes, deterministic code validates — WirelessLife · 2026-11-12