Our comfy-pool image-gen benchmark: z-image vs qwen-image
In short: Our comfy-pool image-gen benchmark: z-image vs qwen-image reports operational numbers measured directly on our ai-server (Hax) stack, and — instead of dumping figures — explains what each number actually means for a real decision. Even if you are new to running local AI, this single post should let you grasp 'what do I decide when I see this number' in
Our comfy-pool image-gen benchmark: z-image vs qwen-image reports operational numbers measured directly on our ai-server (Hax) stack, and — instead of dumping figures — explains what each number actually means for a real decision. Even if you are new to running local AI, this single post should let you grasp 'what do I decide when I see this number' in about five minutes.
| Metric | Measured value | Date | Source |
|---|---|---|---|
| z-image-turbo(8스텝, 1024px, 콜드) 생성 시간 | 6s | 2026-06-30 | Hax ComfyUI 풀 실측 |
| qwen-image(50스텝, 1024px, 콜드) 생성 시간 | 73s | 2026-06-30 | Hax ComfyUI 풀 실측 |
- 표본
- 2 measured metrics (Hax /data curated)
- 측정 환경
- RTX PRO 6000 Blackwell ×4 풀; ComfyUI 0.24.0
- 수집일
- 2026-06-30
- 방법
- 1장 콜드 스타트(모델 로드 포함); 1장 콜드 스타트
What these numbers mean#
같은 1024px 콜드 생성에서 z-image-turbo가 qwen-image보다 약 12.2배 빠르다(6초 vs 73초). 즉, 반복 프리뷰·썸네일엔 z-image로 회전율을 확보하고 최종 품질 컷에만 qwen-image를 쓰는 워크플로 분기가 데이터로 정당화된다.
How we measured it (reproducible conditions)#
These are not vendor specs or marketing figures; they are values we measured ourselves under the conditions below. We list the conditions because when the conditions change, the numbers change too — cold start versus warmed up, batch size, and the exact hardware all shift the result for the same model. So we state reproducible conditions (measured 2026-06-30):
- RTX PRO 6000 Blackwell ×4 풀, ComfyUI 0.24.0, 1장 콜드 스타트(모델 로드 포함)
- RTX PRO 6000 Blackwell ×4 풀, ComfyUI 0.24.0, 1장 콜드 스타트
Rather than memorizing a single number, understand it together with these conditions, so you can diagnose for yourself why your own environment produces a different value.
How to use this in practice#
The derived judgment above translates straight into an operating decision. The point is not to memorize raw figures but to read the relationships between them — a ratio of two values, a utilization rate, a cross-check — because those relationships tell you what to scale up and what to conserve. We use this to check existing headroom before buying new hardware, and to split workflows into a fast path and a quality path. The same logic applies directly to your own local AI setup.
Why this beats vendor specs#
These are values measured in our own operating environment, not vendor sheets or someone else's benchmark. Every number above is measured (not estimated), with date and source (Hax /data). Unlike generic AI-written prose, this derived judgment cannot be produced without the measurement. Only our own measured values are used; no private tokens or internal paths are exposed.
Note: the values above are our own stack measurements as of 2026-06-30 and are refreshed when conditions change (measured values only, no estimates).
Related reading: 로컬 이미지 생성, 2026년 FLUX·SDXL 중 무엇을 골라야 하나?, 로컬 이미지 생성(SDXL·Flux) VRAM·RAM 실측
Responses
No responses yet. Be the first to respond.