➡️ 이 리포는 Motif-3 Beta 기반입니다 — 정식판 리포를 사용하세요 / This repo quantizes the Beta checkpoint — use the official Motif-3 builds
정식 Motif-3 체크포인트 기반 양자화를 게시했습니다. 새로 받으신다면 정식판 리포를 사용해 주세요:
I have published quantizations built from the official Motif-3 checkpoint — please use those instead:
2026-07-22 업로드분에 결함이 있었습니다(discussion #1):
융합 전문가 텐서(gate_up_proj, 4.03e9 요소 = 2³¹의 1.88배)를 나눌 때
mx.split이 4 GiB 오프셋 이후를 침묵 손상시켰고,
이 빌드는 우회 적용 전에 만들어졌습니다. 현재 저장소는 재빌드본이며 정상입니다.
The 2026-07-22 upload was defective (mx.split silent corruption above 2³¹ elements).
This repository now contains a rebuilt, verified model.
검증 / verified: greedy(--temp 0) 5개 슬라이스 전부 정상 — KO 사실·추론·장문, EN, 코드
(distinct-4gram 0.95–1.00, 루프 0건). 구 빌드는 같은 프롬프트에서 무한 반복했습니다.
확증 / forensics: 구·신 빌드의 전문가 0–3번이 바이트 단위 동일 — 손상 오프셋 아래에서는
일치하고 그 위에서만 갈립니다 (손상 시작점: 384개 전문가 중 205번).
Apple Silicon (MLX) 8-bit quantization of
Motif-Technologies/Motif-3-Beta
— a 314.84B-parameter (≈13B active) Mixture-of-Experts reasoning model, text-only,
bilingual Korean / English (+ code), with 97.84% of its mass in the routed experts and
native 256K context. This is the near-lossless reference build: uniform 8-bit
(g64), 8.503 bpw measured, 312 GB on disk, for 512 GB Apple Silicon Macs. Part of
the first community MLX port of Motif-3-Beta in any format.
This build is the anchor of the family: it is statistically near-lossless versus the
bf16 original, and every other build's quality is measured against it (the bf16 model
does not fit in memory, so Q8 is the practical ground truth). Run it when you want the
maximum-quality MLX artifact or a KL/eval reference.
Which Motif-3-Beta MLX build fits your Mac — 2.3bpw / 128 GB / 85 GB, 4.5bpw / 256 GB / 167 GB, 8bit / 512 GB / 312 GB
⚠️ Requires a patched mlx-lm with the motif model class (see How to run). Stock
mlx-lm has no Motif support and will not load this checkpoint.
⚠️ Non-commercial research license, inherited in full from Motif-3-Beta (a beta
release). See License.
Model at a glance
Motif-3-Beta is a text-generation LLM (no vision/audio) built on several non-standard
components. The port reimplements each for MLX and verifies it against a bug-fixed reference.
Motif-3-Beta architecture: embedding to 53 decoder layers (GDLA attention, mHC hyper-connections, dense/MoE feed-forward with Grouped PolyNorm) to RMSNorm to LM head
Sampling: temperature 1.0, top-p 0.95 are recommended (greedy can loop on some
prompts, even at 8-bit — it is a base-model tendency, not a quantization artifact).
Stop tokens: the tokenizer stops on {0, 3, 6}; pass all three as stop ids if you
drive generation yourself.
Reasoning mode: the chat template opens a <think> block automatically — expected
behavior, not a template bug.
Long context: on 512 GB, quantize the KV cache for long prompts
(--kv-bits 8 --kv-group-size 64 --quantized-kv-start 4096). See Hardware.
Method — alis-dwq pipeline
Weights are produced with alis-dwq, the same
pipeline used for the GLM-5.2 (745B) and Hy3 (295B) Alis MLX builds. This build is a
straight uniform 8-bit affine quantization (g64), with embed/head at 6-bit and the
router / mHC / lambda_proj / norms kept fp — no clip-search or DWQ pass. It is the
quasi-continuous source the more aggressive builds are quantized and distilled from, and
the reference they are measured against.
The port itself is parity-verified: MLX logits match a bug-fixed HF reference to
KL ≈1e-7 / token (see Port notes).
Evaluation
This build is the reference — near-lossless versus the bf16 original — so the family's
KL and loop-probe numbers are measured against it, not for it. Fluent bilingual
generation is confirmed: Korean (including hanja), English, and code, with the reasoning
<think> trace intact. General knowledge, code, and bilingual generation are reliable.
The rest of the ladder, measured vs this 8bit reference (fixed EN / code / KO slice):
The 2.3bpw floor build measured against this reference: KL(Q8 ‖ build) and greedy loop-probe across raw / +clip / +clip+DWQ — clip-search + DWQ cut KL 58-72% and break the code loop
Everything in the family is measured against this build. The chart tracks the 2.3bpw floor
build's recovery toward this reference through clip-search and DWQ.
Honest caveat (a base-model property, shown clearly by the reference): very-long-tail
factual recall can be weak — e.g. the model does not reliably reproduce obscure text such as
the 4th verse of the Korean national anthem. Because this build is near-lossless, that
behavior is the base model's, not a quantization artifact; the smaller builds inherit it.
Correctness & port notes
Logit parity. On a truncated model (fp32, EN/KO/code prompts) MLX matches a bug-fixed
HF reference to KL ≈1e-7 / token, top-1 100%. SWA long-context self-consistency is a
bit-exact match (cached wide forward vs no-cache reference).
The shipped HF reference had bugs (documented in the repo discussion
Motif-3-Beta/discussions/6):
a YaRN RoPE dimension crash, an eager-attention GQA crash, and a grouped_mm path that
applied expert-0 PolyNorm coefficients to all experts. Motif Technologies pushed fixes
(SHA d2c9ac6) and this port matches the fixed reference. The root cause of the early
incoherence was the Grouped PolyNorm activation (sigmoid coefficients + output_scale
0.5 + routed bias_clamp) — credit to Motif for the fix.
One MLX-core bug is worked around in the port: mx.split silently corrupts the tail
of a >2³¹-element tensor (the 8 GB bf16 gate_up); basic strided slices are used instead.
Upstream ml-explore/mlx#3836.
Hardware
Built for 512 GB Apple Silicon (M3 Ultra class). Weights are 312 GB, leaving ample room
for KV cache and long-context prefill activation within a 512 GiB machine; use int8 KV
(--kv-bits 8) for the longest prompts. If your Mac has 256 GB, run the
4.5bpw build; for
128 GB, the 2.3bpw floor build.
License
Motif-3-Beta ships no formal license name and no LICENSE file; its model card states,
verbatim:
"Permission is granted to use, modify, and redistribute this software for personal,
educational, and non-commercial research purposes only. Commercial use is prohibited
without prior written permission from Motif Technologies."
This is a non-commercial research license that explicitly permits modification and
redistribution for personal / educational / non-commercial research use. A quantized MLX
port is a modification-and-redistribution, so it is permitted provided it stays
non-commercial and carries the same terms. This derivative inherits that license in
full. The license: other / license_name fields are a best-effort mapping — the
authoritative terms are the quoted text and the base model's card.
Credits & attribution
Base model:Motif Technologies — Motif-3-Beta
(non-commercial research). All architecture — GDLA, Grouped PolyNorm, mHC — and the
trained weights are theirs, as are the reference fixes (SHA d2c9ac6).
MLX port, motif model class, mixed-precision recipe: Alis (avlp12), via
alis-dwq.
Independent, unofficial derivative — not affiliated with or endorsed by Motif Technologies.
Provided as-is, without warranty. Research / non-commercial use only.
Citation
Alis (avlp12) (2026). Motif-3-Beta-Alis-MLX-8bit — near-lossless 8.503 bpw MLX
reference build of Motif-3-Beta
for 512 GB Apple Silicon; the family's KL / eval ground truth.
