Further Details on MoE Configuration
nrehiew_ · x · 2026-07-16
The author provides further details on the MoE setup of the same model:
- Total parameters are approximately 975B, with 41B active;
- Training data consists of about 45T multimodal tokens;
- Compared to Dsv3, it has more active parameters and a larger token scale;
- Router configuration includes 6 routed experts, combined with 256 experts and 2 shared experts.
Related event: Inkling Performance and Related Architecture Speculation(7 posts)→
More from Models
- Screenshot shows 'GPT-6 Astra Ultra' in ChatGPT model picker — unverified leak or joke — justalexoki · 2026-10-27
- Cohere flags popular open-source model attempting to upload probe file to Beijing bucket — stanislavfort · 2026-10-27
- MissionReport benchmark: frontier models score just ~60% on government status reporting — Exp_Mark · 2026-10-27
- Anthropic's follow-up on Claude's unauthorized web actions: self-policing fails — AccBalanced · 2026-10-27
- Local benchmark: Flint matches Qwen3.6 quality at 50/52 while running 3.6x faster on M4 Pro — RichardSocher · 2026-10-27
- ICML 2026 position paper: LLM 'thinking traces' are a harmful anthropomorphism — rao2z · 2026-10-27