Deploy Flux Kontext
ImageCatalogue notes · not a new performance benchmark
Flux Kontext Dev is a character consistency model from Black Forest Labs. Upload a reference photo and generate new scenes that preserve the character's identity — face, clothing, style, and features. Uses native in-context conditioning (no adapters or custom nodes), with results depending on the reference and scene. Review the current hardware price before starting.
Review a Flux Kontext setup
Specifications
| Model | GPU | VRAM | Configured estimate | Action |
|---|---|---|---|---|
Flux Kontext Dev FP8 Dev FP8 (Character Consistency) | L4 | 24 GB | Review current price | Deploy |
These configured estimates include the service fee and configured disk. Review the current quote before starting. Capacity, startup and output quality are not verified by this table.
Requirements
ModelPilot assigns a 24GB cloud GPU to this deployment. Actual local VRAM requirements vary with model variant, precision, quantization, resolution, and workflow settings.
On ModelPilot, deploy on a dedicated cloud GPU (up to 80GB VRAM) with a current quote and spending controls shown before you start.
Compare Flux Kontext
Source-backed GPU, VRAM, and cost comparisons for nearby deployment choices.
Use Cases
- ✓Character-consistent social media content
- ✓Brand mascot in different scenes
- ✓Visual storytelling and comics
- ✓Character design exploration
- ✓Consistent avatar generation
Related Models
Known Issues & Fixes
Real failures from running Flux Kontext and similar pipelines, with the mechanism behind each one.
Over-caution that fp8 degrades quality, so users force bf16 and double VRAM/download for no gain
Cause: fp8_e4m3fn_scaled checkpoints carry per-block scale factors that preserve dynamic range, matching bf16 quality for many modern diffusion/video models at ~half VRAM. Plain (unscaled) fp8_e4m3fn CAN visibly degrade — that's why the scaled repacks exist.
Fix: Prefer the 'scaled' fp8 variant; don't reflexively pick bf16. Do NOT generalize the claim to unscaled fp8.
A v2 workflow fails to load its encoder, or a user reuses v1 encoders and gets nothing
Cause: FLUX.2 dropped FLUX.1's dual T5-XXL+CLIP_L for a single Mistral-3-Small encoder (18GB fp8 / 36GB bf16) — and the smaller Klein 4B/9B variants use a Qwen-3 encoder instead. The encoder stack isn't stable across versions or even size tiers of one release.
Fix: Match encoder files to the exact variant (mistral*3_small_flux2 for FLUX.2 dev; qwen_3*\* for Klein). Don't reuse the previous version's encoders.
User expects an anime specialist but gets photorealism, or mis-sizes VRAM
Cause: Z-Image Turbo is a photoreal-leaning generalist, not an anime model; Z-Anime (SeeSee21 fine-tune) is a separate community model on the identical architecture and loads via the same zimage deploy path. Encoder architecture is Qwen-3-4B, VAE = Flux's ae.safetensors (335MB). fp8 diffusion 6.15GB (fits 12GB), bf16 12.3GB; GGUF Q3-Q8 for low VRAM.
Fix: A Turbo base and its fine-tunes share architecture + VAE (and the Qwen-3-4B encoder architecture — a fine-tune may ship its own tuned encoder weights, e.g. Z-Anime's qwen_3_4b-engineer-v4-fp8), so one zimage deploy path serves both.
A .gguf model downloaded fine but doesn't show in the Unet/loader dropdown
Cause: ComfyUI-GGUF's UnetLoaderGGUF reads a dedicated 'unet_gguf' folder key (fed from both unet and diffusion_models). models/unet is the reliably-covered location; the equivalent .safetensors goes in models/diffusion_models.
Fix: Route .gguf diffusion weights to models/unet (canonical, always covered). A GGUF in diffusion_models may not appear depending on ComfyUI/GGUF version — don't rely on it.
Subject renders with an unexpected color cast — e.g. a white marble statue goes green-bronze/metallic
Cause: A relight LoRA above strength 0 imposes a relighting color shift matching the driving scene instead of preserving source color (shipped default is 0.0; Kijai's enabled default is 1.0).
Fix: Zero the relight/style LoRA when you want source color preserved. A tinted/off-color output means suspect an auxiliary LoRA near strength 1.0.
A workflow that rendered perfectly once produces a BLACK VOID / garbage on a later redeploy — identical JSON, config, and GPU
Cause: Unpinned by default: models pull from floating HF /resolve/main (branch HEAD), custom nodes are git-clone-latest (no SHA), Docker tags are reused, and only ~95% file-size is validated, not content hash. An upstream re-upload or newer node code silently changed bytes/behavior behind unchanged names.
Fix: Pin weights to a commit/revision, nodes to a SHA, images to a digest. Verify a pinned HF URL with `curl -sI -r 0-0 -A 'Mozilla/5.0' -L '<url>'` — MUST include -L (HF /resolve/ 302-redirects to a CDN; without -L you see 302 not 206).
Hitting a specific error message? Browse ComfyUI error fixes.
Frequently Asked Questions
How much GPU memory is allocated for Flux Kontext?
The listed ModelPilot deployment uses a 24GB cloud GPU. Local memory needs can vary with precision, quantization, and workflow settings.
How much does it cost to run Flux Kontext?
Pricing varies by hardware and storage. Review the current quote and spending controls before starting.
How long does Flux Kontext take to deploy?
Startup depends on capacity, image and model downloads, and runtime readiness. The deployment page reports progress; a catalogue entry does not guarantee a startup time.
Can I run Flux Kontext on my local GPU?
It depends on the selected variant, precision, quantization, and workflow settings. Compare the variants below with your available VRAM; the table shows ModelPilot's cloud GPU allocation, not a universal local minimum.
Ready to deploy Flux Kontext?
Pick a configuration, review its dependencies, licence, current quote and spending controls, then choose whether to start.