Duplicate from EllaPriest45/MinimaxH3_Checkpoints

Co-authored-by: Ella G. Priest <EllaPriest45@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,39 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+MiniMax-H3[[:space:]]FL2VA-Curve[[:space:]]Q5_1[[:space:]]-[[:space:]]MinimaxH3.gguf filter=lfs diff=lfs merge=lfs -text
+MiniMax-H3[[:space:]]REF2VA-Curve[[:space:]]Q5_1[[:space:]]-[[:space:]]MinimaxH3[[:space:]]-[[:space:]]0.6str,8steps.gguf filter=lfs diff=lfs merge=lfs -text
+Eros[[:space:]]MAX[[:space:]]beta4[[:space:]]Q5[[:space:]]-[[:space:]]MinimaxH3[[:space:]]-[[:space:]]6-8steps,euler+simple,sampling[[:space:]]shift[[:space:]]12video7audio.gguf filter=lfs diff=lfs merge=lfs -text
+Eros[[:space:]]MAX[[:space:]]beta4[[:space:]]Q5[[:space:]]-[[:space:]]MinimaxH3[[:space:]]-[[:space:]]6-8steps,euler+simple,sampling[[:space:]]shift[[:space:]]12video7audio.mp4 filter=lfs diff=lfs merge=lfs -text

10eros ref2va - MinimaxH3 - fl2va 1str,ref2va 0.7-0.8str,no turbo lora.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5512323817271bde085832ff06e362305cd91cfe9dbf6b59c0c320c34902edaf
+size 13101964656

10eros ref2va - MinimaxH3 - fl2va 1str,ref2va 0.7-0.8str,no turbo lora.txt

@@ -0,0 +1,15 @@
+10eros ref2va - MinimaxH3 - fl2va 1str,ref2va 0.7-0.8str,no turbo lora
+
+Usage.
+
+    Load with LoraLoaderModelOnly (stock ComfyUI, no custom nodes needed).
+
+    On fl2va pruned bf16: strength 1.0 = the complete fine-tune.
+
+    On ref2va pruned bf16: start at 0.7–0.8. The fine-tune was trained in FL2VA context, so at high strength it can compete with reference-image adherence — lower strength trades fine-tune character for reference fidelity. Find your balance.
+
+    Strength scales the whole fine-tune continuously — something the full checkpoint can't do.
+
+    refva turbo lora (4steps) seems to not be enough for good results yet... still testing so not 100% sure. Anyways, safer to use higher steps or simply bypass the turbo lora.
+
+Requirements. A MiniMax H3 pruned-family checkpoint (bf16 recommended; applying onto int8/nvfp4 quants works via runtime patching but bf16 is cleanest), plus the standard H3 stack (Qwen3-VL text encoder, H3 video/audio VAEs).

DaSiWa REF2VA Hybrid INT8 - MinimaxH3 - res_multistep+simple,20-25steps,shift video 10-12,shift audio 3-5.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71c61492faf65b410d0726840ac3b27b017fcfeb76b16ae11589223d81b7121c
+size 20970385305

Eros MAX beta4 Q5 - MinimaxH3 - 6-8steps,euler+simple,sampling shift 12video7audio.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0df4e63f2df63ddaa4d9e7969002a3bc06fadc92620f2ab4b97cf7f1131eabe1
+size 16622013952

Eros MAX beta4 Q5 - MinimaxH3 - 6-8steps,euler+simple,sampling shift 12video7audio.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56418b24ee353858b1b5964b12b9abcc03a3b1c1e1559383b749a0a9f5fb500b
+size 26586874

MiniMax-H3 FL2VA W4E8 - MinimaxH3 - 10steps,544x960,243frames.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18ce05a9e588618db5ae6dca380f242bf8041d9d47f3ffb5cd80d9dec9f9b1ec
+size 12540870488

MiniMax-H3 FL2VA-Curve Q5_1 - MinimaxH3.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6110fe6f6992bfb1ec86518aa92a0644fec4facf25e561430aebfc2a5f4a874
+size 15227425440

MiniMax-H3 FL2VA-Curve Q5_1 - MinimaxH3.txt

@@ -0,0 +1,14 @@
+MiniMax-H3 FL2VA-Curve Q5_1 - MinimaxH3
+
+need ComfyUI 0.30.0 or newer
+
+Before you load one of these: ComfyUI-GGUF does not know this architecture out of the box.
+Its loader keeps a hardcoded allow-list, and minimax_h3 is not in it, so you get ValueError: Unexpected architecture type in GGUF file: 'minimax_h3'. Install ComfyUI-H3-Multishot and run python apply_gguf_arch_patch.py from its folder, then restart ComfyUI. It is one line, idempotent, and applies whichever DiT loader you use.
+
+You also need the text encoder (Qwen3-VL GGUF + its mmproj sidecar) and the VAEs. A file here is the transformer and nothing else — on its own it generates nothing. Load the encoder with the pack's H3ClipLoaderAny at type = minimax, not with “CLIPLoader (GGUF)”, which has no minimax type and fails with a confusing tokenizer error instead.
+
+You also need the text encoder
+
+A file here is the transformer and nothing else. The DiT alone generates nothing. Get the Qwen3-VL encoder (Civitai / Hugging Face) and take the mmproj file with it — it is required for image conditioning and for multi-shot chaining, which feeds the previous shot's last frame through the encoder's vision path. VAEs come from Comfy-Org/MiniMax-H3.
+
+The encoder and the DiT together do not fit on a 32 GB card. Evict the encoder once conditioning is computed — the node pack's keyframe and multishot nodes do this for you, and it is worth roughly a 4× difference in render time.

MiniMax-H3 REF2VA W4E8 - MinimaxH3 - 10steps,544x960,243frames.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddb5e0590ec4702b8024c123c3e8b1d64539ba85446860e86c45783be8c3a09f
+size 12540870488

MiniMax-H3 REF2VA-Curve Q5_1 - MinimaxH3 - 0.6str,8steps.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9449412e88247fd6e6ad6eb2fa9f3dff7b2295b575e70e5c584a14abb4032b85
+size 15227425440

PinkCherry v0.5a I2V FL2VA Pruned INT8 - MinimaxH3.txt

@@ -0,0 +1,33 @@
+PinkCherry v0.5a I2V FL2VA Pruned INT8 - MinimaxH3
+
+PinkCherry MM H3 (Alpha)
+
+Alpha v0.5
+
+** Aug 08 2026 - received some great feedback/comments on weak areas in the checkpoint, and ill work on targeting those weaknesses with further training. thanks for all the input and please keep it comming so I know what you are seeing/hearing.
+
+** Aug 8 2026 - both unpruned (32GB) and pruned (20GB) versions of 0.5-alpha are uploaded. Wont be any new versions of this model for at least a week. as im working on the ref2va model now
+
+Join our discord: https://discord.gg/GCrr4Cj3D
+
+Loras: any lora works with this base. We have had good luck using pussy synth, riding and coach bates penis lora at about str 0.4 to address the faults in this alpha release base.
+
+I2V/T2V: Your biggest win will be with image to video, not text to video at this stage of training this model.
+
+Model Type: First Frame/Last Frame FL2VA , however it worked fine in the reference workflow for me. Ill train ref model later
+
+Model Training: Full fine tune of fl2va base model
+
+Training Dataset: 5000+ 4k videos of various concepts
+
+Model Training Target: FFN/Attn blocks only
+
+Training System: Custom Training Pipeline
+
+    no guardrails or censorship has been bypassed in this model
+
+This is early still in training model I am working on. Treat it as alpha level still. its the first last frame model not the reference model.
+
+Fair Warning: Expect to find dicks on foreheads, exploding breasts and various other issues typical of alpha models. If that's not your thing, wait for beta.
+
+Training Captions used: cock, pussy, vagina, penis, doggystyle, missionary, slut, moaning, orgasm, wet, hairy pussy, spreading, asshole, tight, fucked, riding, cowgirl, anal, reverse cowgirl, pounding, thrusting, hanging tits, fondling, massage, squeezing, dildo, sex toy, penetrate, squats, flexing, bent over, sucking, licking, tongue, blowjob, arousal fluids, jiggling, swaying, sex position, prone, nude, clothing, saggy, breasts, big tits, legs spread, panties pulled to side, cum, semen, creampie, saliva, masturbating, fingers, fingering, eyes closed, perky tits, rubs, lustful, areolas, pink nipples, mouth open, topless, exposed, voyeur, public, bounce, labia, folds, slut, whore, POV, overhead, low angle, high angle, close up, dripping, precum, round ass, amateur, glistening, facing camera, standing sex position, vulva, slutty, tattooed, pounds, from behind, hanging, stroking, jerking, pants down, panties, cock head, cock tip, wet cunt, passionately , rough, gangbang, group fuck, lesbian, threesome, handjob, double blowjob, throbbing, rapidly, rhythmically, wet lips, cleavage, jiggling breasts, cupping, tittyfuck, teasing, erotic, romantic, making out, blindfolded, thick, crotch, deep, steady, aroused, erect, soft flesh, slamming, messy, slapping, pink anus, bathroom

PinkCherry v0.6 INT8 - MinimaxH3.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cb2812f061003d9f345186d58f1bafbf902c6ad2b4c064590b4fc4811634ad1
+size 20970380016

README.md

@@ -0,0 +1,3 @@
+Hi, this whole repository is just a backup of interesting stuff from Civitai in case they pull it all.
+
+All credit to the original artists and lora creators.

RedCraft A2A INT8 - MinimaxH3.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc99ff051283ee05f29b1ebcb14e0d7b36c03e93512ac5479411cdfa2e284122
+size 20970384488