ComfyUI Node

flux1

Flux.1 LoRA training on the cloud — dev or schnell, you pick

By civitai·Created 2 months ago·Updated about a month ago· 42
flux1
  • continue_from
  • api_config
  • moderation_status
  • epochs
  • workflow_id
  • raw_json
model_variantdev
training_data_json
storage_buzz_per_epoch0.00
default_steps0
uses_step_pricingfalse
max_batch_size0
samples_json
epochs1
steps1
batch_size1
lr0.00
text_encoder_lr0.00
train_text_encoderfalse
lr_scheduler
optimizer_type
network_dim1
network_alpha1
noise_offset0.00
flip_augmentationfalse
shuffle_tokensfalse
keep_tokens0
trigger_word

Flux LoRA training is the most-run training job in the whole ecosystem, and this is the node that does it without a local GPU. CivitaiTrainingAiToolkitFlux1 runs Ostris's AI Toolkit trainer on Civitai's cloud to train a LoRA on Flux.1 - you pick dev or schnell, upload a zip of images, and each epoch delivers a downloadable model. No 24GB card, no Kohya environment, no three-hour setup guide. Just Buzz.

It sits in Civitai/Training/flux1, part of Civitai's official ComfyUI pack. The model_variant choice matters: dev is the higher-quality flagship (and the one most style/character LoRAs target), while schnell is the distilled, step-efficient variant - faster to train and fine if your LoRA is for quick drafts. Default is dev, and for most people that's the right call.

How it works

The node submits a training workflow (engine: ai-toolkit, ecosystem: flux1) to Civitai's Orchestration API. Your training images go up as a zip, the cloud runs AI Toolkit (the trainer the community reached for first on Flux), and each epoch produces a checkpoint you can download and drop into a Flux generation workflow. AI Toolkit's tradeoff applies here as it always does: fewer exposed hyperparameters than Kohya, but it just works.

The core input is training_data_json, a JSON object pointing at the hosted zip:

{"type": "zip", "sourceUrl": "urn:air:flux1:dataset:civitai:789@1", "count": 20}

sourceUrl is an AIR URN to the zip of training images; count is the number of images, which is how the API prices the run.

The inputs that matter

  • model_variant (required) - dev (default) or schnell.
  • training_data_json (required) - the zip + count object.
  • epochs / steps - epochs = saved checkpoints (each a downloadable model), steps = total training length and the main pricing control. Set one, the other derives.
  • trigger_word - the token that activates your LoRA in prompts.
  • network_dim / network_alpha, lr, batch_size, noise_offset - the standard trainer dials, all optional. Flux likes network_dim in the 8–32 range for a quick character/style LoRA.

The required-looking storage_buzz_per_epoch, default_steps, uses_step_pricing, max_batch_size fields are generator plumbing - leave them at defaults.

Outputs: moderation_status, epochs, plus the standard workflow_id and raw_json.

Installing it

This is one of ~160 nodes in Civitai Comfy Nodes, Civitai's official pack for their Orchestration API:

  • ComfyUI Manager: Manager → Custom Nodes Manager → search Civitai Comfy Nodes → Install, then restart.
  • CLI: comfy node registry-install civitai-comfy-nodes
  • Source: cd ComfyUI/custom_nodes && git clone https://github.com/civitai/civitai-comfy-nodes.git && pip install -r civitai-comfy-nodes/requirements.txt (just requests).

You need a Civitai account with Buzz and credentials - a Civitai Auth node, CIVITAI_API_TOKEN for headless, or a stored key via the Civitai sidebar.

Where people get burned

  • training_data_json must be valid JSON with type, sourceUrl, and count. Malformed input fails locally before it submits.
  • The pricing is per-step plus per-epoch storage. The cost report on the node after a run is your friend - epochs are where the bill creeps.
  • Moderation gates the model. Your LoRA goes through Civitai's content review, moderation_status reports the verdict, and the platform's real-person rules apply to training data.
  • Flux licensing context. Flux's license has been read as barring NSFW LoRAs, and Civitai removed NSFW Flux models accordingly - another reason to keep training content within platform rules.
  • Early preview. The README warns of breaking changes without notice, and early community reports include bugs and slow jobs. Long trainings can hit the default 30-minute timeout - raise it via the Auth node or CIVITAI_COMFY_TIMEOUT.

If you've got a character or style you want as a Flux LoRA and no free GPU hours, this is the fastest road to it. Feed it a clean, consistent image set and let the cloud do the sweating.

CategoryCivitai/Training/flux1

Inputs (24)

NameTypeDefaultDescription
model_variantCOMBOdev2 options: dev, schnell
training_data_jsonSTRINGRepresents training data in various formats
storage_buzz_per_epochFLOAT0.000–2147483647Per-epoch surcharge (buzz). Each epoch is a delivered checkpoint plus its preview samples, billed on top of the per-step training cost — so raising the epoch count raises the price by this much each. Override per ecosystem where per-epoch samples are expensive to compute (e.g. video).
default_stepsINT00–2147483647Default total step budget when neither Civitai.Orchestration.Grains.Workflows.Steps.Training.AIToolkit.AIToolkitTrainingInput.Steps nor Civitai.Orchestration.Grains.Workflows.Steps.Training.AIToolkit.AIToolkitTrainingInput.Epochs is supplied. Override per ecosystem where the default training length differs (e.g. video needs more steps, quickly-overtrained models need fewer).
uses_step_pricingBOOLEANfalseTrue when billing uses the per-step model. This is the default; the only exception is the legacy path where the caller supplied Civitai.Orchestration.Grains.Workflows.Steps.Training.AIToolkit.AIToolkitTrainingInput.Epochs but no Civitai.Orchestration.Grains.Workflows.Steps.Training.AIToolkit.AIToolkitTrainingInput.Steps (existing consumers), which keeps the historical flat per-epoch price.
max_batch_sizeINT00–2147483647Ecosystem-specific maximum training batch size — the upper bound the user's Civitai.Orchestration.Grains.Workflows.Steps.Training.AIToolkit.AIToolkitTrainingInput.BatchSize is clamped to. Most ecosystems cap at 1.
samples_jsonoptSTRINGSample generation configuration for training workflows
epochsoptINT11–200Number of training epochs — the number of saved checkpoints produced (each epoch yields one downloadable model). When omitted it is derived from Civitai.Orchestration.Grains.Workflows.Steps.Training.AIToolkit.AIToolkitTrainingInput.Steps; when both are supplied, both are honored (epochs = checkpoint count, steps = total).
stepsoptINT11–10000Total number of training steps. This is the primary control over training length and determines pricing. When supplied, Civitai.Orchestration.Grains.Workflows.Steps.Training.AIToolkit.AIToolkitTrainingInput.Epochs (the number of saved checkpoints) is derived from it; when omitted, steps are derived from epochs.
batch_sizeoptINT11–4Training batch size. Defaults to 1; raise it (up to the ecosystem's maximum) to train faster at the cost of more GPU memory. A larger batch sees more images per step, so fewer steps are needed for a comparable result. Values above the ecosystem maximum are clamped down.
lroptFLOAT0.000–1Sets the learning rate for the model. This is the learning rate when performing additional learning on each attention block (and other blocks depending on the setting).
text_encoder_lroptFLOAT0.000–1Sets the learning rate for the text encoder. Only used when TrainTextEncoder is true. For models with multiple text encoders, this applies to all of them.
train_text_encoderoptBOOLEANfalseWhether to train the text encoder(s) alongside the model. Enabling this can improve prompt understanding but increases training time and memory usage.
lr_scheduleroptCOMBOYou can change the learning rate in the middle of learning. A scheduler is a setting for how to change the learning rate.
optimizer_typeoptCOMBOThe optimizer determines how to update the neural net weights during training. Various methods have been proposed for smart learning, but the most commonly used in LoRA learning is "adamw8bit".
network_dimoptINT11–256The larger the Dim setting, the more learning information can be stored, but the possibility of learning unnecessary information other than the learning target increases. A larger Dim also increases LoRA file size.
network_alphaoptINT11–256The smaller the Network alpha value, the larger the stored LoRA neural net weights. For example, with an Alpha of 16 and a Dim of 32, the strength of the weight used is 16/32 = 0.5, meaning that the learning rate is only half as powerful as the Learning Rate setting. If Alpha and Dim are the same number, the strength used will be 1 and will have no effect on the learning rate.
noise_offsetoptFLOAT0.000–1Adds noise to training images. 0 adds no noise at all. A value of 1 adds strong noise.
flip_augmentationoptBOOLEANfalseIf this option is turned on, the image will be horizontally flipped randomly. It can learn left and right angles, which is useful when you want to learn symmetrical people and objects.
shuffle_tokensoptBOOLEANfalseRandomly changes the order of your tags during training. The intent of shuffling is to improve learning. If you are using captions (sentences), this option has no meaning.
keep_tokensoptINT00–10If your training images have tags, you can randomly shuffle them. However, if you have words that you want to keep at the beginning, you can use this option to specify "Keep the first 0 words at the beginning". This option does nothing if the Shuffle Tokens option is off.
trigger_wordoptSTRINGA trigger word that activates the trained LoRA when used in prompts. Only applicable to certain ecosystems (sd1, sdxl, flux1, chroma, zimagebase, zimageturbo, flux2klein).
continue_fromoptCIVITAI_AIROptional previously-trained LoRA to continue training from ("train further"). When set, the first epoch resumes from this model instead of the base model, and the new epochs build on top of it.
api_configoptCIVITAI_CONFIGOptional Civitai Auth connection; defaults to CIVITAI_API_TOKEN or stored OAuth login.

Outputs (4)

NameTypeDescription
moderation_statusSTRING
epochsSTRING
workflow_idSTRING
raw_jsonSTRING