Reset Flux 3 Klein T2I results

This commit is contained in:
Morpheus
2026-10-28 17:07:46 +00:00
parent f5d3cddb8a
commit 4caccae494
18 changed files with 57 additions and 2594 deletions
+33 -73
View File
@@ -1,84 +1,44 @@
# FLUX 3 Image / Klein
This is the canonical local FLUX.3 Klein text-to-image workflow and test
matrix. New reference-based tests belong in `optimized/I2I/flux3klein/` and
`optimized/MultipleRef2I/flux3klein/`; do not create a second T2I directory.
The canonical cross-model comparison row is `1088×1920`; the other recorded
resolution rows are explicitly labelled scaling studies.
This is the canonical local FLUX.3 Klein text-to-image workflow and clean
sampler/scheduler test plan. Reference-based tests belong in
`optimized/I2I/flux3klein/` and `optimized/MultipleRef2I/flux3klein/`.
This directory contains the first controlled workflow and test plan for the
server's locally installed FLUX 3 Image Klein model. The workflow does not use
No results have been retained for this fresh run. The workflow does not use
the remote `Flux3ImageNode` API.
## Workflow
## Fixed comparison inputs
- `workflow.json` — API-format prompt for `/prompt` automation.
- `prompt.txt` — fixed baseline prompt.
- `variables.json` — fixed values and allowed test variables.
- `results/` — output images and matching metadata records.
- Workflow: `workflow.json`
- Prompt: `prompt.txt`
- Model: `flux3-image-klein-int8-convrot.safetensors`
- Text encoder: `qwen3vl_4b_int8_convrot.safetensors` (`type=flux3`)
- VAE: `flux2-vae.safetensors`
- Resolution: **1088×1920**
- References: none
- Seed: `310001`
- Steps: `4`
- Flux Guidance: `3.5`
The local graph uses `flux3-image-klein-int8-convrot.safetensors`,
`qwen3vl_4b_int8_convrot.safetensors` with `type=flux3`, `flux2-vae.safetensors`,
Flux guidance, Euler sampling, and the native ComfyUI sampler chain. The
remote API node is not used because it requires a separate Comfy API login.
The graph uses `FluxGuidance`, not `CFGGuider`; therefore this T2I suite
records Flux Guidance rather than calling it CFG. Prompt, resolution, seed,
steps, guidance, and reference count stay fixed while sampler and scheduler
vary.
## Test matrix
## Fresh test matrix
The first pass changes one variable at a time. The prompt, aspect ratio,
grounding state, safety tolerance, and reference count remain fixed unless a
row explicitly changes them.
The matrix contains **48 samplers × 9 schedulers = 432 planned runs**. All
rows are currently **Planned**. The complete sampler and scheduler lists are
stored in `variables.json`; each executed cell must receive a result record in
`results/` with its exact settings, runtime, output path, dimensions, and
error or visual notes.
| ID | Purpose | Resolution | References | Grounding | Seed | Status | Runtime (s) | Verdict |
|---|---|---:|---:|---|---:|---|---:|---|
| F3I-T2I-01 | Shared-prompt local baseline | 1080×1920 target | 0 | Off | 310001 | **Passed** | 19.6 | 1088×1920 output generated |
| F3I-T2I-02 | Lower-cost portrait comparison | 0.75K target | 0 | Off | 310002 | **Passed** | 5.9 | 768×1360 output generated |
| F3I-T2I-03 | Detail/quality portrait comparison | 1.5K target | 0 | Off | 310003 | **Passed** | 13.9 | 1536×2720 output generated |
| F3I-T2I-04 | Higher-resolution portrait feasibility | 2K target | 0 | Off | 310004 | **Passed** | 47.2 | 2048×3648 output generated |
| F3I-T2I-05 | Grounded prompt comparison | 1K | 0 | On | 310005 | **N/A local** | — | Grounding belongs to remote API node |
| F3I-I2I-01 | Single-reference edit | 1K | 1 | Off | 310006 | **Passed** | 33.6 | 1024×1024 output generated |
| Matrix | Resolution | References | Seed | Guidance | Steps | Samplers | Schedulers | Status |
|---|---:|---:|---:|---:|---:|---:|---:|---|
| F3I-MATRIX | 1088×1920 | 0 | 310001 | 3.5 | 4 | 48 | 9 | Planned (0/432) |
## Evaluation
Record the exact prompt, seed, resolution, reference filename/hash, runtime,
output dimensions, success/failure, and visual notes in `results/<test-id>.json`.
Keep the first successful baseline as the comparison target. Do not promote a
larger resolution solely because it is larger; judge detail, composition,
prompt adherence, artifacts, and cost/runtime together.
The API node's seed controls rerun identity but the service may remain
nondeterministic. That is recorded as a property of the test, not treated as a
failure.
## Matrix results
The initial resolution and reference runs were calibration tests with a short
fox prompt. They remain useful for checking graph/model compatibility but are
superseded for model comparison by the shared prompt baseline. The native
reference conditioning path passed with `Base001.png`. The grounding row is not
run locally because grounding is a capability of the authenticated remote
`Flux3ImageNode`, which is intentionally excluded from this repository's
workflow.
## Initial run
The earlier API-node probe is retained as `results/F3I-T2I-01-remote-api.json`
as an environment diagnostic: it returned `Unauthorized` before generation.
The local workflow was then corrected to use the Flux 3-compatible
Qwen3-VL 4B encoder and 128-channel `flux2-vae.safetensors`; that run passed
and is recorded in `results/F3I-T2I-01.json`.
## Full sampler/scheduler matrix
The local model was tested with every available sampler and scheduler: 48 × 9
combinations, for 432 total runs. All 432 completed successfully at 1088 ×
1920, four steps, seed 310001, using the shared prompt. The complete cell-level
results, runtimes, and server output prefixes are recorded in
[`results/F3I-MATRIX.json`](results/F3I-MATRIX.json). The analysis and visual
spot-check findings are in [`MATRIX-ANALYSIS.md`](MATRIX-ANALYSIS.md).
The recommended baseline remains **Euler + Simple**. `dpm_adaptive` is not
recommended for routine use because it was an extreme runtime outlier; runtime
success does not by itself establish visual quality.
The five strongest visually reviewed outputs are collected in
[`results/best5/`](results/best5/), ordered best first.
Do not silently substitute a rejected sampler or scheduler. Record failures
as results, preserve the fixed comparison inputs, and update this table only
from the actual result records. The first successful `euler + simple` cell is
the compatibility baseline; no visual winner should be declared until the
outputs have been reviewed under the same prompt and resolution.