Share prompt across T2I workflows
This commit is contained in:
@@ -4,4 +4,6 @@ Use [`flux/`](flux/) or [`qwen/`](qwen/) for the best tested text-to-image basel
|
||||
|
||||
Static variables are recorded in [`variables.json`](variables.json). The test plans live in each model README.
|
||||
|
||||
The shared comparison prompt is [`qwen_prompt.txt`](qwen_prompt.txt); all T2I model workflows should use this prompt so model results remain directly comparable.
|
||||
|
||||
Store each T2I test under the selected model's `results/<test-id>.png` and `results/<test-id>.json` using the table ID as the label.
|
||||
|
||||
@@ -21,6 +21,8 @@ Files:
|
||||
|
||||
The included seed and prompt are a controlled baseline example. For reuse, replace the prompt, seed, and resolution; sampler, scheduler, steps, and CFG are the optimization controls.
|
||||
|
||||
The positive prompt is the shared T2I comparison prompt in [`../qwen_prompt.txt`](../qwen_prompt.txt), also used by the Qwen workflow.
|
||||
|
||||
## Planned optimization matrix
|
||||
|
||||
All rows are intentionally **Untested**. Keep the prompt, seed, and resolution from [`../variables.json`](../variables.json) fixed. Flux2Scheduler does not expose a selectable scheduler, so that value remains fixed.
|
||||
|
||||
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
@@ -1,7 +1,7 @@
|
||||
{
|
||||
"test_id": "Q-T2I-01",
|
||||
"workflow": "T2I/qwen",
|
||||
"prompt_file": "../qwen_prompt.txt",
|
||||
"prompt_file": "../../qwen_prompt.txt",
|
||||
"prompt_sha256": "5802909ebe894260b70a3977430e08aaa95732e0559749d8cb42e2f5ecf3ab6f",
|
||||
"negative_prompt_source": "workflow.json node 5 negative_prompt",
|
||||
"seed": 20260920,
|
||||
|
||||
Reference in New Issue
Block a user