Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-09-24 · source-backed
500 reasoning and 500 knowledge questions after a year of cleanup, evaluated at high reasoning with and without tools, with a recommended tool-use setup on GitHub (lastexam.ai). Per-model scores exist only in chart images, so I haven't verified the exact numbers. Reception reads as a reset after labs saturated the original.
Each link below shares sources, entities, or timing with this story.
The announcement describes the same underlying model at two safeguard levels: Fable generally available, Mythos restricted to vetted cybersecurity and life-sciences organizations, currently US-only. Fable 5.1 scores 52.6% on Terminal-Bench-Science 0.1 (against 24.7% for Fable...
The abliteration tool gained 215 stars to reach 30,103, but the stronger signal is downstream: the HF trending endpoint returns DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NM-DAU and Momoking/Qwen3-VL-32B-Heretic-MiniMax-H3-NVFP4, both naming the too...
Announced August 13, up to 14x Standard processing, 11x faster than Claude Fable 5, 5x faster than Opus 4.8 on Fast mode, running on the Cerebras Wafer-Scale Engine with 44 GB of on-chip SRAM per wafer. Limited API preview for select customers with capacity-gated expansion. Th...
The study names it Solution Hacking: reaching the right answer through numerical search, enumeration, guessing, or answer-first verification rather than a valid derivation. It scales with difficulty, 2.2% on common problems, 28.3% on Olympiad-level, 37.4% on Humanity's Last Ex...
Published July 30: 276B total parameters with only 12B active, 1M-token context, open weights on Hugging Face. 31.6% on Humanity's Last Exam, 80.2% on SWE-Bench Verified, 82.2% on IFBench. Natively multimodal across text, image and audio, variable reasoning effort, $1.20 per 1...
BAAI's AREX (24 authors, 124 upvotes on HF Daily Papers) alternates between gathering evidence and drafting provisional answers, then audits those answers constraint-by-constraint. The distinguishing mechanism is a learned autonomous context-update tool that compresses growing...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.