Gemini 4 Pro Argon Checkpoint Rumored Ghost-Testing on Arena, GPT-6 Sol Enters Grey Testing
WorldofAI · youtube · 2026-09-18
This AI news roundup aggregates several unverified leaks and updates:
- Gemini 4 Pro leaks: the rumored Argon checkpoint is reportedly being ghost-tested on Arena, with impressive coding, gaming, and SVG outputs.
- GPT-6 Sol has entered grey testing, with early comparisons against GPT-5.6 Sol and Astra.
- Grok 4.7 update and new benchmark results.
- Google DeepMind's Dream-RSI research, plus a reveal that Union Alpha is Pareto 26.9.
- Xiaomi's large-scale MiMo-V2.6 RL run and Claude's new Projects workflow.
All leaks are third-party and unconfirmed. The video includes sponsor and community promotions.
More from Models
- Skeptical deep dive confirms Humanity's Last Exam errors; official o3-mini grader marked right answers wrong every time — paul_cal · 2026-09-20
- Matt Shumer asks if Jev could help with scalable oversight and alignment checks — mattshumer_ · 2026-09-20
- FrontierSWE v2 opens 24.1-point gap: Claude Fable 5.1 scores 56.29% vs GPT-5.6's 32.2% — geoffwolfe · 2026-09-20
- 22M local model beats JEV 93% vs 80% on Banking77 in 8ms on CPU — Prompt Engineering · 2026-09-20
- Jev loses to Gemini on 1,565-email classification benchmark, but dev still wants it in production — socialwithaayan · 2026-09-20
- Jev Detector scans ~10,000 words for AI slop in ~2 seconds, free with no sign-up — socialwithaayan · 2026-09-20