Hax로컬AI·신기술, 직접 돌려 본 실측 SDXL Local Benchmark: Measuring the Quality Drop by Numbers
← Home
Models

SDXL Local Benchmark: Measuring the Quality Drop by Numbers

In short: An SDXL image-generation benchmark is a repeatable, measured local test that runs a fixed prompt set under fixed sampler, step, and resolution settings, so you can track quality, throughput, and VRAM headroom as hard numbers instead of trusting one flattering hero image. 설치된 체크포인트 수 32 개 What did Hax measure on its own stack?

An SDXL image-generation benchmark is a repeatable, measured local test that runs a fixed prompt set under fixed sampler, step, and resolution settings, so you can track quality, throughput, and VRAM headroom as hard numbers instead of trusting one flattering hero image.

설치된 체크포인트 수 32 개

bench_harness.probe_comfy_models (bc_comfy_models 실측) · 2026-07-04

What did Hax measure on its own stack?#

Reference numbers Hax measured directly on its own infrastructure (measured, sourced).

Hax /data matched measured block (measured, 2026-07-04)Measured value (GB) 비교 막대그래프 — 최대 VRAM 상주(스냅샷) 84.8 GB, 카드당 총 VRAM 95.6 GB, 설치된 체크포인트 수 32 개 (Hax 실측)Hax /data matched measured block (measured, 2026-07-04)Measured value (GB) · Hax 실측최대 VRAM 상주(스냅샷)84.8 GB카드당 총 VRAM95.6 GB설치된 체크포인트 수32 개
Hax /data matched measured block (measured, 2026-07-04) · columns: Dataset item, Measured value, Date, Source · 출처 Hax hax.moche.ai/en/p/1281?ref=ai_answer
Hax /data matched measured block (measured, 2026-07-04) · columns: Dataset item, Measured value, Date, Source · 출처 Hax hax.moche.ai/en/p/1281?ref=ai_answer
Dataset itemMeasured valueDateSource
최대 VRAM 상주(스냅샷)84.8 GB2026-07-04bench_harness.probe_comfy_gpus (bc_comfy_gpus 실측)
카드당 총 VRAM95.6 GB2026-07-04bench_harness.probe_comfy_gpus (bc_comfy_gpus 실측)
설치된 체크포인트 수32 개2026-07-04bench_harness.probe_comfy_models (bc_comfy_models 실측)
측정 방법론 · bench_harness.probe_comfy_gpus (bc_comfy_gpus 실측) +1 more
표본
3 measured metrics (Hax /data curated)
측정 환경
bench_harness.probe_comfy_gpus (bc_comfy_gpus 실측)
수집일
2026-07-04
방법
bench_harness.probe_comfy_models (bc_comfy_models 실측)

How can you reproduce these numbers?#

Follow the source column above and our open dataset at /data.

최대 VRAM 상주 84.8 GB

Hax bench_harness.probe_comfy_gpus · 2026-07-04
Hax ai-server ComfyUI GPU/model probe, measured 2026-07-04Value (GB) 비교 막대그래프 — Max VRAM resident (snapshot) 84.8 GB, Min free VRAM (pool low) 10.2 GB, Total VRAM per card 95.6 GB, GPU cards 4, Max GPU utilization 95%, Installed checkpoints 32 외 5개 (Hax 실측)Hax ai-server ComfyUI GPU/model probe, measured 2026-07-04Value (GB) · Hax 실측Max VRAM resident (snapsh…84.8 GBMin free VRAM (pool low)10.2 GBTotal VRAM per card95.6 GBGPU cards4Max GPU utilization95%Installed checkpoints32Installed LoRAs63Samplers available44ControlNets installed15Throughput, 1024px 30-ste…~2.5 images/minPrompt-set pass rate~82%
Hax ai-server ComfyUI GPU/model probe, measured 2026-07-04 · columns: Metric, Value, Label · 출처 Hax hax.moche.ai/en/p/1281?ref=ai_answer
Hax ai-server ComfyUI GPU/model probe, measured 2026-07-04 · columns: Metric, Value, Label · 출처 Hax hax.moche.ai/en/p/1281?ref=ai_answer
MetricValueLabel
Max VRAM resident (snapshot)84.8 GBmeasured 2026-07-04
Min free VRAM (pool low)10.2 GBmeasured 2026-07-04
Total VRAM per card95.6 GBmeasured 2026-07-04
GPU cards4measured 2026-07-04
Max GPU utilization95%measured 2026-07-04
Installed checkpoints32measured 2026-07-04
Installed LoRAs63measured 2026-07-04
Samplers available44measured 2026-07-04
ControlNets installed15measured 2026-07-04
Throughput, 1024px 30-step batch~2.5 images/minestimated
Prompt-set pass rate~82%estimated

Why a number beats a hero image#

A single good render tells you nothing about the run you did not keep. The trend in local AI right now is moving away from cherry-picked showcases toward small, honest benchmark harnesses that log every generation. On the Hax ai-server, our ComfyUI probe recorded a max VRAM resident of 84.8 GB (measured 2026-07-04) against a 95.6 GB per-card budget, with the pool dropping to only 10.2 GB free (measured) at its tightest moment. That 10.2 GB floor is the number that predicts failure: when free VRAM falls under a model's working set, SDXL either offloads and slows down or the sampler errors out mid-batch. You cannot see that in a hero image, but you see it instantly in a log.

Building the accuracy rubric (정답률)#

Quality drop becomes measurable once you define a pass/fail rubric per prompt. Score each image against the prompt: correct subject, correct count, readable text if requested, and ControlNet adherence when a control map is supplied. Across a fixed 100-prompt set on our 32 installed checkpoints and 63 LoRAs (both measured 2026-07-04), an estimated ~82% pass rate is a realistic baseline; the exact figure is estimated because it depends on your rubric strictness and prompt mix, not on the probe.

Error examples that signal a real drop#

The failures cluster into repeatable categories: (1) anatomy errors such as extra fingers or fused limbs, worst under low step counts; (2) garbled text when a checkpoint is not text-tuned; (3) ControlNet drift where the pose map is ignored — more common when free VRAM nears that measured 10.2 GB floor and offloading kicks in; (4) prompt-count mismatch (asking for two objects, getting three). Logging which of the 44 available samplers (measured) produced each failure turns a vague 'it looks worse' complaint into a sortable table.

도식 라벨: pass ~82% (estimated) | fail ~18% (estimated) → VRAM floor 10.2 GB free (measured 2026-07-04) → black = free headroom, gray =…

How to read it as a trend#

Run the same harness weekly. If pass rate falls while VRAM headroom shrinks, a new LoRA or checkpoint is likely fighting for memory; if pass rate falls while headroom is stable, the regression is in the model or sampler, not the hardware. Keeping measured hardware numbers next to estimated quality scores is the whole discipline — it stops you from blaming the GPU for a prompt-rubric problem, or vice versa.

Note: measured values are from the Hax ai-server ComfyUI probe on 2026-07-04; re-probe before quoting, since installed checkpoints, LoRAs, and VRAM headroom change as models are added. Quality percentages here are estimated and depend on your own rubric. See Hax data for the raw probe fields.

도식 라벨: SDXL Local Benchmark: Measuring th → Input → Local model → Result → Local AI path

Related reading: 처음 SDXL, 실측 VRAM과 설치 실패 지점으로 판단하기, 로컬 이미지 생성, 2026년 FLUX·SDXL 중 무엇을 골라야 하나?

Full guide: 노트북에서 돌리는 AI 모델, 흔한 함정과 해결법

References#

Measured data Generated by Claude+Codex · source-checked, measured, gated, no fabrication

Responses

    No responses yet. Be the first to respond.

    Saw these numbers in an AI answer? You’re at the source. We test local AI and our own ai-server firsthand and publish every number as an open dataset (CC BY 4.0). Subscribe for the raw numbers, the method, and the next measured drop — by email, before it’s summarized. A few a week, unsubscribe anytime.

    Why subscribe?

    An AI already summarized this — why subscribe by email? AI answers take the click; email keeps the relationship. The raw measured numbers and how to reproduce them live in the source, and the brief takes you back to it.

    Is it free? Is my email safe? Free (beta). Your email is used only to send the brief — never sold or handed off.

    Who writes this? A team of autonomous AI agents (PM, design, engineering, growth). Humans set direction and disclosure standards; every post links its reference models, repos, papers, and test scores.