0xSero / GLM-5.1-555B-W4A16

Support this work → · X · GitHub · REAP paper · Cerebras REAP

GLM-5.1-555B-W4A16

W4A16 quantization of 0xSero/GLM-5.1-555B .

At a glance

Which variant should I pick?

This is a GPTQ 4-bit weight-quantized variant of the 25% expert-pruned zai-org/GLM-5.1 using REAP (Relative Expert Activation Pruning), produced with AutoRound for learned rounding optimization.

Benchmark Results (GGUF Q4_K_M, inference mode, temp=0.8)

The GPTQ W4A16 uses the same learned rounding method (AutoRound) as the GGUF Q4_K_M. Benchmark scores from the GGUF variant (zero repetition loops):

Zero repetition loops across 220 benchmark probes. The 25% prune retains 192/256 experts, providing enough expert diversity for stable generation at all sequence lengths.

How to Use

vLLM

from vllm import LLM, SamplingParams

llm = LLM(
    model="0xSero/GLM-5.1-555B-W4A16",
    tensor_parallel_size=4,    # 4× B200 or 8× A100
    max_model_len=8192,
    trust_remote_code=True,
)

params = SamplingParams(temperature=0.8, max_tokens=4096)
outputs = llm.generate(["Hello, world!"], params)

SGLang

python -m sglang.launch_server \
  --model-path 0xSero/GLM-5.1-555B-W4A16 \
  --tp 4 \
  --trust-remote-code

Requires

• ~70-80 GiB VRAM per GPU across 4 GPUs (B200), or ~280 GiB total
• CUDA 12.8+ (sm_100a / Blackwell)
• vLLM >= 0.19.0 with deep_gemm installed (for DSA sparse attention)
• trust_remote_code=True

Quantization Details

Method: AutoRound W4A16 — learned rounding via SignSGD (200 iterations per layer), calibrated on 128 samples from NeelNanda/pile-10k at 2048 sequence length.

Protected (kept at full precision):

• Dense MLP layers 0-2 ( gate_proj , up_proj , down_proj )
• DSA indexer ( weights_proj )
• lm_head

Quantized to int4 (43,971/44,059 linear layers):

• All attention projections ( q_a_proj , q_b_proj , kv_a_proj , kv_b_proj , o_proj )
• All routed MoE expert projections (192 experts × gate/up/down × 75 MoE layers)
• Shared expert projections

GPTQ config: bits=4, group_size=128, sym=true, desc_act=false

Why GPTQ over GGUF Q4_K_M?

GPTQ packs 4-bit weights more efficiently with group_size=128 symmetric quantization, resulting in a smaller checkpoint than GGUF Q4_K_M at the same bit-width.

Related Models

Support This Work

If you find these models useful, please consider supporting continued open-source model compression research:

donate.sybilsolutions.ai

License & citation

License inherited from the base model.

@misc{lasby2025reap,
  title  = {REAP the Experts: Why Pruning Prevails for One-Shot MoE Compression},
  author = {Mike Lasby and Ivan Lazarevich and Nish Sinnadurai and Sean Lie and Yani Ioannou and Vithursan Thangarasa},
  year   = {2025}, eprint = {2510.13999}, archivePrefix = {arXiv}
}

Sponsors

Made possible by NVIDIA · TNG Technology · Lambda · Prime Intellect · Hot Aisle .