Skip to main content
ModelPilot

LTX-2 "Value not in list: gemma_path" — the gated Gemma encoder + tokenizer fix

Also seen as: "Value not in list: gemma_path", "gemma_path not in []", "LTXVGemmaCLIPModelLoader value not in list", "gemma-3 tokenizer.model 403"

Direct answer

How do I fix "LTX-2 gemma_path not in list (gated Gemma encoder)"?

LTX-2 (19B) uses a Gemma-3 LLM as its text encoder, loaded via LTXVGemmaCLIPModelLoader as a directory — and it shows an empty gemma_path list when that directory is missing the gated config + tokenizer files. The diffusion weights come from the open Comfy-Org/ltx-2 repo, but config.json and tokenizer.model come from the gated google/gemma-3-12b-it repo. Without an HF token + accepted license, that sidecar 403s and the loader has nothing to list.

Key facts

  • iYou downloaded the LTX-2 weights but not the Gemma encoder’s config.json + tokenizer.model, which live in the gated google/gemma-3-12b-it repo.
  • iNo HF token (or a token without access to that repo), so the gated sidecar files 403.
  • iYou have not accepted the Gemma license on HuggingFace.

Next step

  1. 1Accept the google/gemma-3-12b-it license on HuggingFace while logged in.
  2. 2Create a read token and use it for the download so config.json + tokenizer.model land inside the Gemma encoder directory.
  3. 3Confirm the encoder is a directory (weights + config + tokenizer together) — LTXVGemmaCLIPModelLoader loads a folder, not a single file.
  4. 4On LTX-2.3, switch to the default single-file FP4 encoder (gemma_3_12B_it_fp4_mixed) — it is non-gated and needs no tokenizer sidecar, avoiding the gate entirely.

If that didn't fix it

  • A token alone is not enough — you must click "accept" on the specific gated Gemma repo first, or the sidecar still 403s.
  • This is LTX-2 19B specifically; LTX-2.3’s default single-file FP4 encoder does not hit it.

Not sure this workflow will run before you rent a GPU?

Paste your ComfyUI workflow JSON into the free checker — it flags known dependencies and an estimated hardware requirement. It does not run the workflow, verify GPU capacity or judge output quality. No signup.

Check my workflow freeSee GPU pricing

Other ComfyUI errors

Fix

GPU Memory Exhausted

Fix

Model Not Found

Fix

Missing Python Package

Fix

Disk Space Exhausted

Fix

Gated Model Access

Fix

missing custom nodes (node type not found)

Frequently Asked Questions

What causes "LTX-2 gemma_path not in list (gated Gemma encoder)" in ComfyUI?

You downloaded the LTX-2 weights but not the Gemma encoder’s config.json + tokenizer.model, which live in the gated google/gemma-3-12b-it repo.

How do I fix "LTX-2 gemma_path not in list (gated Gemma encoder)"?

Accept the google/gemma-3-12b-it license on HuggingFace while logged in. Create a read token and use it for the download so config.json + tokenizer.model land inside the Gemma encoder directory.