CLIP input is invalid: None — trace where the text encoder should come from
A wire reaching CLIPTextEncode does not prove that a valid encoder was loaded. Check upstream loading and whether the model uses a complete checkpoint or separate components.
Symptoms and scope
A wire reaching CLIPTextEncode does not prove that a valid encoder was loaded. Check upstream loading and whether the model uses a complete checkpoint or separate components.
Error fragments for search; IDs, values and filenames may vary:
CLIP input is invalid: None
does not contain a valid CLIP
Source-supported context
ComfyUI v0.37.0 raises this error in CLIPTextEncode when its clip input is None. The checkpoint warning in that message applies only when the input came from a checkpoint loader; issue #6819 records one such report, not a universal diagnosis. Basic SD checkpoints can supply CLIP, while split FLUX examples load text encoders separately. Source 1 Source 2 Source 3 Source 4
Cases to distinguish
1. Diffusion-only weights are used in a checkpoint path that expects an included text encoder.
2. The encoder failed, was bypassed or returned no object, with the error surfacing downstream.
3. The encoder architecture or loading mode does not match the model family.
Diagnostic sequence
The sequence below is an editorial procedure based on the cited context, not a diagnosis already confirmed for your environment.
Step 1. Trace the failing CLIP input upstream and record the supplying node class and exact file.
Step 2. Compare the author's example to determine whether it expects a bundled checkpoint or separate diffusion, encoder and VAE files.
Step 3. Supply the correct encoder and loading mode for a split workflow. Do not attach an unrelated SDXL or SD1.5 CLIP merely to fill the socket.
Step 4. Remove optional LoRA, IPAdapter and control branches to isolate the native encoder path. Verify the correct loader produces a CLIP object before changing prompt text; prompt wording cannot repair a None input.
Completion check
The encoder supplies a valid nonempty object and both encoding and sampling advance. Replacing this with a matrix-shape error is not completion.
Limitations and cautions
CLIP is often an interface object name, not a guarantee of one internal encoder architecture. Deleting the prompt-encoding node cannot replace required conditioning.
Original sources
- ComfyUI v0.37.0 CLIPTextEncode implementation — checked 2026-09-25.
- CLIP input is invalid None issue — original record checked 2026-09-21; rechecked 2026-09-25 as an individual report.
- Text-to-image workflow — original record checked 2026-09-21; rechecked 2026-09-25.
- FLUX.1 native workflows — original record checked 2026-09-21; rechecked 2026-09-25.
Sources and node code rechecked on 2026-09-25. The 2026-09-21 dates above record the earlier source review. No GPU run, minimum-memory measurement or runtime guarantee is supplied.
Related troubleshooting and guides
Was this page helpful?
Anonymous. Only a yes/no counter is stored; no account, IP address or device details.
Sources & references
Official text-to-image and FLUX documentation, the v0.37.0 CLIPTextEncode code, and the individual issue were rechecked on 2026-09-25. No GPU run, minimum-memory measurement or runtime guarantee is supplied.
01CLIP input is invalid NoneSource checked: 2026-09-2502Text-to-image workflowSource checked: 2026-09-2503FLUX.1 native workflowsSource checked: 2026-09-2504ComfyUI v0.37.0 CLIPTextEncode implementationSource checked: 2026-09-25