Liquid d1 3B — English Memory Gate

A local decision model designed as an upstream filter for Hindsight and similar conversational memory systems. It returns IGNORE for clearly disposable interaction and CONSIDER for personal information, unresolved references or memory-related requests. CONSIDER means assessment, not automatic storage or permission to execute a command.

Validation

On 57 English test examples from the 296-example curated/synthetic dataset, the selected training adapter scored 56/57 (98.2%) and its F16 llama.cpp export scored 57/57 (100%). One prediction differed between runtimes. These consumed experimental test results are not independent real-conversation validation or a safety guarantee. End-to-end Hindsight integration has not been validated.

License

LFM Open License v1.0 applies; see LICENSE, including the $10 million annual revenue commercial-use threshold. This is not an Apache or MIT license. The base LiquidAI/d1-3B model was modified through English memory-gate LoRA fine-tuning.

This repository: complete merged model

The base and trained adapter are merged into the safetensors checkpoint. No separate adapter is required or should be applied. All adapter layers were removed, the native D1 loader/configuration are included, and saved weights were reloaded with exact tensor equality. This verifies serialization, not a separate accuracy benchmark for this merged-HF runtime.

For llama.cpp deployment, use the separate F16 GGUF repository. The final dataset is English Memory Gate.