GLM-5.3-Flash-VQ-3.6bpw

134.0 GiB — q4-class quality at 32 GiB less than affine q4.

A data-free vector-quantized build of GLM-5.3-Flash for Apple Silicon, fitted from the bf16 checkpoint (598.5 GiB) with no calibration corpus. Built with VQLab.

MoE experts uniformly at d=4 / K=8192 (13-bit packed codes). Attention, embeddings and the output head stay at 8-bit affine; norms, routers and the full 347-tensor vision tower stay bf16. GLM's MTP layer (layers.45) is never quantized.

Instead of rounding each weight onto a uniform grid the way affine quantization does, VQ stores small groups of weights as indices into codebooks fitted to the weights themselves — which is why it beats affine at matched bytes below 6 bits. The method and results are in our paper, Data-Free Vector Quantization Beats Affine Quantization at Matched Bytes Below 6 Bits (CC BY 4.0).

The affine builds compared against below are our own conversions of the same base, made with the same tooling and scored on the same instrument.

Requirements

The glm5_next architecture ships in released mlx-vlm 0.6.17 — a normal pip install, not a fork. The VQ runtime needs no patches: it ships inside the checkpoint as model.py, declared via model_file in config.json, and resolves under both mlx-lm and mlx_vlm.

pip install 'mlx-vlm>=0.6.17'

Vision-ready: config.json carries vision_config and image_token_id, and the vision tower ships bf16. To serve across several Macs, see Run it.

Changelog

2026-10-01 — runtime refresh; Knurlogic

The bundled model.py is updated to the current VQ runtime; weights are unchanged. The run, multi-Mac, drafting and vision instructions now use Knurlogic.

2026-09-09 — runtime refresh

Runtime refresh. The bundled model.py is updated so downloaders run exactly the code that was benchmarked; dense bundles now carry both runtimes.

• Faster prefill on affected geometries, measured per rung: a device-codebook kernel arm for large-codebook geometries (up to 1.46x on affected rungs), a ragged-subvector relaxation (up to 1.34x on affected rungs), a fused d8 arm, and a routing memo (≈2%). No blanket speedup is claimed across the lineup — gains apply only where the geometry engages the new paths.
• Output quality is unchanged: the kernel changes are bit-identical or 1-ULP-equivalent, and the routing memo is bit-identical (logits checksum verified).
• Speculative decoding (MTP): on repos that ship mtp-head-q6.safetensors, the sidecar works with the exo fork branch mtp-stage1 (github.com/noahzelezny/exo) — launch each node with exo --mtp (or set EXO_MTP=1). Note: with MTP enabled, exo serves requests sequentially (the batch engine has no MTP path), so leave it off for concurrent / multi-agent workloads.

Measured results

Referee: 2048 tokens; prose = WikiText, code = public mlx corpus (pinned manifest), literary = Gutenberg. KL is against the bf16 teacher's cached top-64 logits (captured mass 0.9906 on every row).

The frontier claim this rung supports: at 32 GiB less than affine q4 (134.0 vs 166 GiB), it is better on KL (94.54 vs 98.34), better on top-1 agreement, and better on code perplexity, essentially tied on prose and literary. If you were going to run q4, this is the same quality class in substantially less memory.

Rank these by KL, not perplexity. Perplexity is an aggregate over finite text and absorbs offsetting errors; KL measures distance to the teacher's distribution directly.

These perplexity scores aren't comparable across model families. The bf16 teacher has near-verbatim memorized the public corpora used here (mean top-1 probability 0.857 on prose, measured), so absolute perplexity for this family is contamination-dominated. KL to that teacher stays fully valid — a sharp teacher is harder to track.

Runtime

Speed on this revision (measured 2026-10-01)

Split across two Macs with Knurlogic (M3 Ultra 96 GB + M4 Max 128 GB, pipeline over Thunderbolt), against GLM-5.3-Flash-4bit on the same two Macs: 3 pairs, a prompt of about 1,900 tokens, 128 generated tokens, no drafting.

A two-Mac split's absolute speed depends on the link, so only the ratio is quoted.

This rung exceeds any single machine we gate on (134 GiB of weights against a 128 GB box), so it was verified on a 2-node exo pipeline (M3 Ultra 96 GB + M4 128 GB, Thunderbolt): the release gate's cluster smoke generated coherent tokens through this exact artifact, with the peer rank's copy identity-checked (runtime/config/index hashes and every shard size) before the generation counted. Single-box throughput numbers do not exist for this rung and none are quoted; the bundled runtime is byte-identical to the one gated and generation-verified single-box on the 2.7bpw rung.

Memory

• Download: 134.0 GiB (19 shards); repo total including the sidecar is ≈140.1 GiB. Expect resident memory near the download size once routing has touched all experts. This is a 2-node rung, or a single 192 GB-class Mac.
• Keep prefill bounded. The bundled runtime caps MLX's buffer-reuse cache (4 GiB default, VQLAB_CACHE_LIMIT_GB to override, =0 disables), which is what keeps long-prompt peaks near resident size.
• If you run your own serving loop: chunk the prefill (2048) and call mx.eval([c.state for c in cache]) plus mx.clear_cache() after every chunk — the chunk size only bounds the peak if each chunk is actually forced.
• Single-box peak memory is not measured for this rung: it does not fit one machine. Cluster figures are above.

Run it

Requires a Mac with Apple Silicon (macOS, MLX).

With Knurlogic, which works out the settings and checks the model fits before loading it:

pip install knurlogic
hf download TheDrainFlorist/GLM-5.3-Flash-VQ-3.6bpw --local-dir ~/Knurlogic/Models/GLM-5.3-Flash-VQ-3.6bpw
knurlogic serve ~/Knurlogic/Models/GLM-5.3-Flash-VQ-3.6bpw

Chat at http://127.0.0.1:8080/, or point any OpenAI-, Anthropic- or Ollama-compatible client at it (model local). knurlogic ui opens a page with a Launch button for every model on disk instead.

At 152 GB on disk this needs a Mac with more memory than that, or two Macs: install the same Knurlogic version and the model on each, run knurlogic ui --host cluster on both, and launch it from the page with both machines selected. Knurlogic splits the layers between them.

Images work through Knurlogic directly; mlx-vlm is not needed.

The VQ runtime ships inside this repo as model.py (declared by model_file in config.json), and Knurlogic runs the file the model ships.

Without Knurlogic:

python -m mlx_vlm generate \
  --model TheDrainFlorist/GLM-5.3-Flash-VQ-3.6bpw \
  --prompt "Explain vector quantization briefly." \
  --max-tokens 512

Speculative decoding (MTP)

mtp-head-q6.safetensors (6.09 GiB) is GLM's own multi-token-prediction head — layers.45, 889 tensors, packed to q6. The head is shared across the GLM-5.3-Flash-VQ lineup because that layer is never quantized in any rung; it was validated end-to-end on the 2.7bpw rung (acceptance 0.827 over greedy decoding, output distribution exactly the base model's via rejection sampling). It is never named in the weight index, so stock loaders ignore it entirely; nothing changes unless you opt in.

Cluster speculative decoding is live for this rung. Knurlogic loads the shipped head and drafts automatically when the model is split across Macs; there is nothing to enable. The trunk verifies every drafted token by exact rejection sampling, so outputs are exactly the base model's. Validated on this exact artifact on a 2-node exo pipeline; acceptance 0.70–0.87 depending on content. Throughput varies with shard placement on mixed-generation clusters (chip generations trade compute against bandwidth), so benchmark your own topology. The head adds ≈6.3 GiB on the rank that draws it.

On a single box the MTP head roughly breaks even against plain decode (measured on the 2.7bpw rung: 19.99 vs 19.7 tok/s); its payoff is pipelined serving, where drafting hides inter-stage latency.

The mtp-head-q6.safetensors files in our lineups (Qwen Flash 2.14 GiB, Qwen 397B 5.41 GiB, GLM 6.09 GiB) are different heads with different geometry across families and share only a filename. Never cross-copy them between families.

Methodology

Fitted data-free from the bf16 checkpoint — k-means / Lloyd over weight subvectors, seed 1234, no Hessian, no activation statistics, no calibration corpus. All 42 expert layers (L3–L44) at uniform d=4/K=8192; codes are packed sub-byte into uint32 words (13 bits per code) with an fp16 scale per (row, 64 weights). The seed-noise floor for this family's geometry is 6.32 mnats on KL, well below every gap in the table above.

Verification

Release gates passed on this artifact before upload: file, index and tokenizer checks, a verbatim match between the bundled runtime and its source, and a generation smoke through the shipping runtime on Apple Silicon. The upload path runs the gate itself and refuses to publish without it.

Limitations

• Verified on a 2-node cluster, not single-box. Generation on this exact artifact ran through an exo pipeline; no single-box throughput or peak-memory figures exist yet.
• Perplexity scores aren't comparable across model families — see the note under Measured results.
• MTP throughput on clusters is placement-sensitive (see the MTP section); single-box, this rung does not fit one machine.

Paper

The method, the full model ladder, the negative results, and the measurement rules behind every number here: Data-Free Vector Quantization Beats Affine Quantization at Matched Bytes Below 6 Bits (CC BY 4.0) · code: VQLab · web version: Space

Support this work

VQLab and these artifacts are built and released independently — the fits, the measurement harness and the published rungs are one person's compute and time. If they are useful to you:

• Sponsor: github.com/sponsors/noahzelezny
• Contract work: available for quantization and on-device inference work on Apple silicon — custom rungs, per-layer allocation for your model, or getting a checkpoint to run well on a Mac. Contact: hello@thedrainflorist.com

Provenance

Base model: zai-org/GLM-5.3-Flash — MIT licensed. This is a quantized derivative and inherits that license; using it means accepting the base model's terms. Quantization: TheDrainFlorist, 2026.

The upstream authors ask that their technical report be cited in research use:

@misc{glm5team2026glm5vibecodingagentic,
      title={GLM-5: from Vibe Coding to Agentic Engineering},
      author={GLM-5-Team and : and Aohan Zeng and Xin Lv and Zhenyu Hou and Zhengxiao Du and Qinkai Zheng and Bin Chen and Da Yin and Chendi Ge and Chenghua Huang and Chengxing Xie and Chenzheng Zhu and Congfeng Yin and Cunxiang Wang and Gengzheng Pan and Hao Zeng and Haoke Zhang and Haoran Wang and Huilong Chen and Jiajie Zhang and Jian Jiao and Jiaqi Guo and Jingsen Wang and Jingzhao Du and Jinzhu Wu and Kedong Wang and Lei Li and Lin Fan and Lucen Zhong and Mingdao Liu and Mingming Zhao and Pengfan Du and Qian Dong and Rui Lu and Shuang-Li and Shulin Cao and Song Liu and Ting Jiang and Xiaodong Chen and Xiaohan Zhang and Xuancheng Huang and Xuezhen Dong and Yabo Xu and Yao Wei and Yifan An and Yilin Niu and Yitong Zhu and Yuanhao Wen and Yukuo Cen and Yushi Bai and Zhongpei Qiao and Zihan Wang and Zikang Wang and Zilin Zhu and Ziqiang Liu and Zixuan Li and Bojie Wang and Bosi Wen and Can Huang and Changpeng Cai and Chao Yu and Chen Li and Chengwei Hu and Chenhui Zhang and Dan Zhang and Daoyan Lin and Dayong Yang and Di Wang and Ding Ai and Erle Zhu and Fangzhou Yi and Feiyu Chen and Guohong Wen and Hailong Sun and Haisha Zhao and Haiyi Hu and Hanchen Zhang and Hanrui Liu and Hanyu Zhang and Hao Peng and Hao Tai and Haobo Zhang and He Liu and Hongwei Wang and Hongxi Yan and Hongyu Ge and Huan Liu and Huanpeng Chu and Jia'ni Zhao and Jiachen Wang and Jiajing Zhao and Jiamin Ren and Jiapeng Wang and Jiaxin Zhang and Jiayi Gui and Jiayue Zhao and Jijie Li and Jing An and Jing Li and Jingwei Yuan and Jinhua Du and Jinxin Liu and Junkai Zhi and Junwen Duan and Kaiyue Zhou and Kangjian Wei and Ke Wang and Keyun Luo and Laiqiang Zhang and Leigang Sha and Liang Xu and Lindong Wu and Lintao Ding and Lu Chen and Minghao Li and Nianyi Lin and Pan Ta and Qiang Zou and Rongjun Song and Ruiqi Yang and Shangqing Tu and Shangtong Yang and Shaoxiang Wu and Shengyan Zhang and Shijie Li and Shuang Li and Shuyi Fan and Wei Qin and Wei Tian and Weining Zhang and Wenbo Yu and Wenjie Liang and Xiang Kuang and Xiangmeng Cheng and Xiangyang Li and Xiaoquan Yan and Xiaowei Hu and Xiaoying Ling and Xing Fan and Xingye Xia and Xinyuan Zhang and Xinze Zhang and Xirui Pan and Xu Zou and Xunkai Zhang and Yadi Liu and Yandong Wu and Yanfu Li and Yidong Wang and Yifan Zhu and Yijun Tan and Yilin Zhou and Yiming Pan and Ying Zhang and Yinpei Su and Yipeng Geng and Yong Yan and Yonglin Tan and Yuean Bi and Yuhan Shen and Yuhao Yang and Yujiang Li and Yunan Liu and Yunqing Wang and Yuntao Li and Yurong Wu and Yutao Zhang and Yuxi Duan and Yuxuan Zhang and Zezhen Liu and Zhengtao Jiang and Zhenhe Yan and Zheyu Zhang and Zhixiang Wei and Zhuo Chen and Zhuoer Feng and Zijun Yao and Ziwei Chai and Ziyuan Wang and Zuzhou Zhang and Bin Xu and Minlie Huang and Hongning Wang and Juanzi Li and Yuxiao Dong and Jie Tang},
      year={2026},
      eprint={2602.15763},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2602.15763},
}

Built with MLX and VQLab.