Fetching from the wire…
Public story · 2026-02-20 · source-backed
GitHub | 196B total / 11B active, Apache 2.0
StepFun's sparse MoE activates only 11B of 196B parameters per token, delivering 74.4% SWE-bench, 97.3% AIME 2025, and 100-300 tok/s throughput. Supports INT4 GGUF for local inference. Apache 2.0 licensed. One of the most capable fully open-source agentic models available. Trending on HN with 224 points.
Each link below shares sources, entities, or timing with this story.
Shared entities / Same source domain / Shared topic / What happened next
Both cover Apache, GitHub, MoE, SWE; reported by the same outlet (github.com); overlapping topics (local, model).
Both cover Apache, GitHub, MoE, SWE; reported by the same outlet (github.com); overlapping topics (apache, model).
Shared entities / Shared topic / What happened next / Tension
Both cover AIME, Flash, MoE, SWE; overlapping topics (aime, model); picks up the AIME thread on 2026-06-07.
Google released Flash / Shared entities / Shared topic / What happened next
Linked by a graph relationship (Google released Flash); both cover Apache, MoE; overlapping topics (active, apache, inference, local, model).
Shared entities / Same source domain / Shared topic / What happened next
Both cover AIME, Apache, MoE; reported by the same outlet (github.com); overlapping topics (apache, inference, model).
Google released Flash / Shared entities / Same source domain / Earlier coverage
Linked by a graph relationship (Google released Flash); both cover Apache, GitHub, Trending; reported by the same outlet (github.com).
Shared entities / Shared topic / What happened next
Both cover Apache, Flash, MoE, SWE; overlapping topics (apache, inference); picks up the Apache thread on 2026-04-24.
Both cover Apache, Flash, MoE; overlapping topics (apache, inference, local, model); picks up the Apache thread on 2026-08-10.