Today for AI
HOT RADAR
llmHEAT 9.0°

AI CURATED HOT STORY · 10/20/2026

MemT 1.0 Released: Native Memory Architecture Scales to 300B Parameters

1 reports archived1 independent sourcesupdated 10/20/2026, 12:06:00
Synthesis & Latest Updates
1 Sources Cross-Validated

MemTensor has open-sourced the MemT 1.0 series, marking the first engineering validation of a 'memory-native' architecture at nearly 300B parameters. By embedding parallel memory pathways within Transformers, it encodes experiences as learnable internal dynamic states, addressing long-term memory and knowledge retention challenges in AI Self-Improvement (RSI).

LATEST/MemTensor has open-sourced the MemT 1.0 series, marking the first engineering validation of a 'memory-native' architecture at nearly 300B parameters. By embedding parallel memory pathways within Transformers, it encodes experiences as learnable internal dynamic states, addressing long-term memory and knowledge retention challenges in AI Self-Improvement (RSI).

HEAT TRENDHourly heat curve

3 fully observed hours
Now
9.0
Peak
9.522:00
24h change
–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

TIMELINECoverage timeline

Total 1 reports · Latest first
  1. 量子位 (微信公众号)T2·78 pts
    • MemT 1.0 features Nano (36B) and Flash (288B) versions, built on Qwen3.6 and DeepSeek-V4 bases respectively.
    • The core innovation is a parallel memory computation pathway alongside attention, enabling models to learn writing, updating, and reading experiences directly.
    • Compared to external memory solutions using vector databases, native memory architectures offer higher scalability ceilings and tighter reasoning coupling.
MemT 1.0 Released: Native Memory Architecture Scales to 300B Parameters | Today for AI