Qwen3-VL-32B H3 ComfyUI Generation Tails

This repository contains generation-only tails for Qwen3-VL-32B H3

conditioning encoders. Each tail supplies language layers 50–63, the final

language normalization layer, and the LM head.

These files are not standalone CLIP checkpoints. They reuse the tokenizer,

token embedding, vision tower, and language layers 0–49 from a connected H3

conditioning encoder.

Files

| --- | --- | --- | ---: |

Use a tail from the same source family as the connected 0–49 encoder when

possible. The Instruct tails pair with the H3 encoder derived from

Qwen3-VL-32B-Instruct, including its

NVFP4/AWQ build. The Ultra Heretic tails pair with the Ultra Heretic H3

encoder.

Standalone text and vision-language generation

Install

ethanfel/ComfyUI-H3-Qwen3VL-TextGen

to reconnect an H3 conditioning encoder and one of these tails as a

general-purpose local Qwen3-VL text and vision-language generator.

Load CLIP (H3 0–49 encoder) ── clip ──────────┐
                                              ├─ H3 Qwen VL Generate Text (Standalone)
H3 Qwen VL Generation Tail Loader ─ tail_clip ┘
Optional IMAGE batch ───────────── image ─────┘

The standalone node supports editable system and user prompts, optional image

batches, deterministic or sampled decoding, and Qwen thinking mode. The

connected base CLIP is preserved; only the temporary tail is explicitly

unloaded after generation.

Installation

Place the selected tail under ComfyUI's text-encoder model directory, for

example:

ComfyUI/models/text_encoders/H3/generation_tails/

The filename must contain generation_tail_50_63 so the dedicated loader can

discover it. Restart or refresh ComfyUI after installing a new file.

Then:

1. Load the H3 0–49 conditioning encoder with ComfyUI's standard Load CLIP

node and its H3-compatible encoder type.

2. Select the tail with H3 Qwen VL Generation Tail Loader.

3. Connect clip and tail_clip to **H3 Qwen VL Generate Text

(Standalone)**.

The same tail artifacts can also be used by a compatible H3 prompt-enhancer

workflow.

Formats

BF16

The BF16 tails retain all 156 source tensors at full precision.

INT8 ConvRot

The INT8 tails contain 98 learned row-wise ConvRot transformer matrices and a

chunk-compatible INT8 LM head. Norms and other protected tensors remain BF16.

The ConvRot group size is 256.

NVFP4/AWQ

The NVFP4/AWQ tail contains 98 native ComfyUI NVFP4/AWQ transformer matrices.

Its final norm and 151,936 × 5,120 LM head remain BF16 so the tail loader can

evaluate the vocabulary head through its ordinary full-precision path. The

file contains 576 tensors and is 5.03 GiB.

Runtime verification

The NVFP4/AWQ tail was tested with its matching 0–49 NVFP4/AWQ encoder in a

current ComfyUI container on an NVIDIA RTX PRO 6000 Blackwell. A generation

step traversed all 64 language layers, the temporary tail unloaded cleanly,

the connected base remained at exactly 50 layers, and the returned base CLIP

then produced finite (1, 7, 5120) H3 conditioning.

The BF16 and INT8 tail paths were also exercised through all 64 language

layers and verified to restore the connected 50-layer conditioning encoder

after generation.

Provenance

Ultra Heretic tails:

repository: llmfan46/Qwen3-VL-32B-Instruct-ultra-uncensored-heretic
revision:   c44b949b30d111666a5ed9851c5cd633ed39b070

Official BF16 and INT8 tails:

repository: MiniMaxAI/MiniMax-H3
revision:   83db0c0efe6ef9824e0e194be110346c0a9542ed
architecture: Qwen3-VL-32B-Instruct

Official NVFP4/AWQ tail donor:

repository: cybermotaz/Qwen3-VL-32B-Instruct-NVFP4
revision:   dfcab1395094f56d3c78ba472158284def0fc278

See SHA256SUMS for every file's Hugging Face LFS checksum.