woodfireind/MiniMax-H3-GGUF-MiniStack

MiniMax H3 GGUF MiniStack

A shrunk, GGUF-quantized MiniMax H3 FL2VA stack built for ComfyUI on consumer GPUs (developed on Intel Arc A770 16 GB). The full 33B stack has been reduced via AdaLN pruning of the DiT and distillation of the 32B text encoder down to a 4B student, then quantized to Q4_K_M. The audio VAE is dropped (see Limitations).

Provenance: this is a derived, quantized repack of MiniMaxAI/MiniMax-H3 as repackaged by Comfy-Org/MiniMax-H3, plus locally trained adapter weights. It is released under the MiniMax H3 Community License Agreement.

What's in the box

📂 ComfyUI/models/
├── 📂 diffusion_models/
│   └── MiniMax-H3-FL2VA-pruned-Q4_K_M.gguf       11.38 GB  pruned AdaLN DiT, Q4_K_M
├── 📂 text_encoders/
│   ├── qwen3vl-4b-h3student-Q4_K_M.gguf           2.50 GB  distilled TE, Q4_K_M (default)
│   ├── qwen3vl-4b-h3student-BF16.gguf             8.05 GB  distilled TE, BF16 (optional)
│   ├── te_adapter_v1.safetensors                    120 MB  student→H3 5120-d MLP adapter (required)
│   └── h3_tokenizer/                                  11 MB  H3 tokenizer (required)
└── 📂 vae/
    └── minimax_h3_video_vae_fp16.safetensors       4.85 GB  stock video VAE, fp16

Total download ≈ 27 GB (Q4 default ≈ 19 GB if you skip the BF16 TE).

File	SHA-256
diffusion_models/MiniMax-H3-FL2VA-pruned-Q4_K_M.gguf	51783047a6be43c9fcedc3c0a1746480dac867839ba1ad2e1c698da9d7058201
text_encoders/qwen3vl-4b-h3student-Q4_K_M.gguf	7ff66b1541723fc4c4c3f2cba75f72740fad43276ab1d6d564379e591a2e74e5
text_encoders/qwen3vl-4b-h3student-BF16.gguf	b6c55a51fe20aca2700fcedd228ad9af9b67f896a10ae5ff80092752c09145ba
text_encoders/te_adapter_v1.safetensors	ef2060e1a277820416d895812dd580234d3c2760af8e6d203b38e3df2e0952b0
text_encoders/h3_tokenizer/tokenizer.json	a5d85b6dcc535e6b93115a9ef287e6132fdbf30270da6218194ba742261173c7
text_encoders/h3_tokenizer/tokenizer_config.json	a07e942ac874baa13758de8d1fbdb186683cc03416b5589e1b6671c6b3057c68
text_encoders/h3_tokenizer/vocab.json	ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910
text_encoders/h3_tokenizer/merges.txt	599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3
vae/minimax_h3_video_vae_fp16.safetensors	7c1f131492e7eddacaac9069a61b81bdd39de5cc96561e677c5eab1cdce5e522

Documented changes vs. stock

1. Pruned DiT (FL2VA). The 33B full-width DiT is replaced with the AdaLN-curve form from Comfy-Org/MiniMax-H3 (minimax_h3_fl2va_pruned_bf16): AdaLN 13B→44M params via the adaln_t_table [1025,8] + per-block [96768,8] projections. Effective ≈ 18.7B params.
2. Q4_K_M quantization. Converted with a two-pass streaming converter and quantized with a ComfyUI-GGUF-patched llama.cpp build (arch "wan"; stock llama-quantize rejects this arch). 11.38 GB vs 14.5 GiB for the full Q3_K_M. GGUF: general.architecture=wan, 532 tensors, 53 fp32/fp16 hiprec islands (adaln*, t_embed, norm, patch_proj) kept as-is.
3. Distilled student text encoder. The 32B Qwen3-VL-32B teacher is replaced with a Qwen3-VL-4B student plus a trained MLP adapter (2560→4096→5120, fp32) that maps student language hidden states into H3's 5120-d multimodal token slots. Distilled from 2,500 teacher dumps; val_cos 0.9310 (full-corpus v4 adapter). Q4_K_M GGUF is numerically near-identical to the BF16 student (>0.99 cosine); the gap to the teacher is adapter fidelity, not quant noise.
4. Audio VAE dropped. The minimax_h3_audio_vae_fp32 component is removed from the workflow. Output videos are silent. The stock minimax_h3_video_vae_fp16 is unchanged and included.
5. Video VAE unchanged. Stock fp16 video VAE; no pruning or requantization was applied.

Requirements

• ComfyUI with MiniMax H3 support (Comfy-Org/ComfyUI PR #15224 / current main) and the ComfyUI-GGUF custom node.
• For the student TE you need the h3_small_te custom node (H3SmallTELoader / H3SmallTextEncoder) — it loads the GGUF via ComfyUI-GGUF's gguf_clip_loader into comfy-native Llama2_(Qwen3VL_4BConfig) with GGMLOps, keeping the H3 HF tokenizer + Identity final-norm + the adapter. The adapter and h3_tokenizer/ are required — without them the student cannot encode.
• DiT and video VAE are loaded with the standard UnetLoaderGGUF / VAELoader nodes.

Recommended settings (640×480 previz, Arc A770 16 GB)

• Sampling: 8 steps with the turbo LoRA (larryvrh/MiniMax-H3-Turbo-Lora), --reserve-vram 3. Note the turbo LoRA's adaln_proj targets dangle on the pruned checkpoint — apply it to the full checkpoint if you need those.
• Length: snap to 5 mod 17 frames (17k+5 grid). Example: length = max(5, round(seconds*24)) + (5 - (max(5, round(seconds*24)) % 17)) % 17.
• Default: 640×480 / 124 frames / 24 fps / 8 steps.

Limitations

• No audio. The audio VAE is dropped from this stack — generated videos are silent and there is no voice/SFX/music output.
• Student TE is text-only. The 4B student adapter was distilled on pure-text teacher dumps; it has no vision path. For I2V / FL2V keyframe identity (which needs the teacher's vision tower + deepstack features), use the teacher 32B TE (qwen3vl-32B-MiniMax-H3-*); the student cannot consume image tokens.
• Adapter fidelity ceiling. val_cos ≈ 0.93 vs the teacher. Shot structure and cuts are preserved, but cut timing can drift ±2 s and fine prompt specifics (small props, signage details) can wash out. Timestamp-magnitude tuning (v3/v4) plateaued; this is accepted as a known tradeoff.
• Turbo LoRA compatibility. The preview turbo LoRA targets full-width adaln_proj weights that do not exist in the pruned curve-form checkpoint; using both together silently skips those targets.
• GGUF arch. The DiT is archived as "wan" for quantizer compatibility; use a ComfyUI-GGUF build that can load arch-wan GGUFs. The student GGUF is arch qwen3vl.
• Frames-vs-seconds prompting. H3 is trained on mm:ss.mmm timestamps; the 4× temporal VAE grid only quantizes where cuts can land — frame-based timestamps are off-distribution.
• Not for 2K. Local gen is 640×480 previz; 768P/2K regeneration requires the official MiniMax cloud API.

Usage

1. Unzip files into the ComfyUI models/ folders as shown above.
2. For T2V (text only): UnetLoaderGGUF → pruned Q4 GGUF; H3SmallTELoader → student Q4 GGUF + adapter + tokenizer dir; VAELoader → video VAE.
3. For I2V / FL2V: swap the student TE for the teacher 32B TE (vision path); everything else stays.
4. Encode with MiniMaxH3ImageToVideo, sample with SamplerCustomAdvanced (res_multistep / simple, 25 steps, or 8 with turbo LoRA).

Eval notes (summary)

• Baseline (full Q3_K_M, 25 steps): 620 s wall @ 640×480/124f.
• Pruned Q4_K_M (25 steps): 610 s wall, output on-par with baseline (same composition/cut).
• Turbo 4/8-step: ~160 s / ~250 s wall; 8-step recommended.
• FL2V smoke (pruned Q4 + teacher TE): 295.5 s wall; keyframe lock first/last pixel-cos 0.9905/0.9987.
• Student vs teacher numeric A/B: mean cosine 0.847 (mid adapter) → 0.9310 (full-corpus v4). Video A/B: structure and cuts preserved; fine details degrade consistent with ~0.93 cos.

License

Derived from MiniMaxAI/MiniMax-H3 — MiniMax H3 Community License Agreement (see link in metadata header). The student adapter is a training-derived artifact of distillation against the teacher model and is distributed under the same agreement. The video VAE is the unmodified Comfy-Org repack.