# Heimdall D1 V6 decision model by Toby-NoSkillSon > Heimdall D1 V6 adapts d1-2B to judge who transcribed speech addresses. Read its local MLX formats, training lineage and unmeasured accuracy. Updated 2026-11-07 (Australia/Adelaide) ## Quick answers ### What is Heimdall D1 V6 by Toby-NoSkillSon? [Heimdall D1 V6](): A text-only d1-2B fine-tune that decides whether transcribed speech is for a voice assistant, still unfinished or meant for someone else. [Heimdall D1 V6 facts]() [Source]() ### Who makes Heimdall D1 V6? [Heimdall D1 V6](): Toby-NoSkillSon [Heimdall D1 V6 facts]() [Source]() By Toby-NoSkillSon · Hugging Face package release: 2026-11-04 · OPEN A text-only d1-2B fine-tune that decides whether transcribed speech is for a voice assistant, still unfinished or meant for someone else. It runs through MLX on Apple Silicon. The fixed question was trained around the names Toby and Heimdall; the state can supply other names. The weights preceded this scan window; the in-window card explains their lineage. ## Facts - Inputs: Conversation transcripts; speech recognition runs first - Formats: BF16, int8 and int4 packages - Licence: LFM Open License v1.0; commercial revenue limit applies ## Claims and sources - **VENDOR-REPORTED**: The maker reports two training stages totalling about 175 million tokens across ten languages. The card gives no held-out dispatch accuracy, so training scale does not establish reliable wake-word-free behaviour. [source]() ## Links - [Source: Card, weights and MLX runner]() ## Related projects No related projects are listed in the registry.