GLM-4.5-Air · vLLM · (quant TBD)

status: blocked

Configuration

Modelzai-org/GLM-4.5-Air
CompanyZhipu AI
FamilyGLM
Parameters106B / 12B (MoE)
EnginevLLM
Quant / precisionTBD
Why this quantA human should pick the quant — the two fitting options trade official-but-tight against community-but-comfortable (see Notes).
Downloadzai-org/GLM-4.5-Air
Context window65536
Input modalitiestext

Measured results

Prefill tok/s
Decode tok/s
Peak memory (GB)

Full run command

# blocked — pick a quant first (see Notes), then fill in and run.

glm-4.5-air Zhipu AI GLM 41-130B conc-32

Notes

Blocked — needs a human to specify the quant before running. Added 2026-06-22.

GLM-4.5-Air (106B total / 12B active MoE) is the only GLM that realistically fits a single Spark for the text benchmark — the bigger GLMs are out of reach (GLM-4.6 is a ~355B MoE → too large even at 4-bit; GLM-4.6V is vision-first → wrong methodology, like phi-4-reasoning-vision). It’s the natural substitute for the blocked glm-4-7-flash slot. But the two fitting quants trade off in ways that warrant a human call:

Quant Size Fit on 121 GB Source Caveat
zai-org/GLM-4.5-Air-FP8 ~106 GB tight — needs --gpu-memory-utilization ~0.95 + modest ctx, may OOM official (Zhipu, compressed-tensors) quality-safe, but little KV headroom
cpatonn/GLM-4.5-Air-AWQ-4bit ~55 GB comfortable community (cpatonn) not a top-trusted quantizer (same tier concern as the Devstral AWQ)

Decision needed: which quant to benchmark —

  • FP8 (official, accept the tight-fit / OOM risk and reduced context), or
  • AWQ-4bit (comfortable fit, but accept the community-quantizer source per the “trusted repo” policy), or
  • wait for an official 4-bit (e.g. a RedHatAI quantized.w4a16 — none exists yet).

Once a quant is chosen, set quant/source_repo/download_url, add the quant tag, flip to pending, and run via scripts/bench-vllm-serving.sh.