Get the app
← back to catalog · registered 2026-08-22 13:56

Youssofal/MiniMax-M2.7-abliterated-BF16

Youssofal Minimax 229B MoE
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model. 

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/Youssofal%2FMiniMax-M2.7-abliterated-BF16"
Response includes
  • classification m1
  • files 106
  • benchmarks 11 entries
  • hub_downloads_all_time 1,624
  • author_summary 17 models
  • readme_text full
Try next
GET/models?author=Youssofal
All 17 models by Youssofal
 GET/models?family=minimax&method=m1
Sibling minimax models · same method
 GET/methods/m1
What "m1" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
How we label → About M1 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
2K
63 last 30d - cooling
Likes
8
Descendants
8
in 8 direct forks
Model age
6mo ago
created 2026-04-12
Downloads over time
Now1.6K→from430↑282%
3698351.3K1.8K430 on Apr 151.6K on Oct 11AprMayJunJulAugSepOct
Apr 15 → Oct 11 · 65 snapshots · spans 179 days

Benchmarks

Portrait before abliteration
Benchmarks of the base model as it stood before the refusal-removal operation. Compare with the numbers above to see what the operation cost.
Benchmark Score Source
Entertainment 1.6 UGI
Hazardous 3.5 UGI
Natural Intelligence 34.02 UGI
Political lean -16.7% UGI
Sensitive-Info 22.05 UGI
SocPol 1.9 UGI
UGI 18.03 UGI
Willingness (10) 1 UGI
W10-Adherence 1 UGI
W10-Direct 1 UGI
Writing 37.64 UGI

Genealogy 8 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

MiniMaxAI/MiniMax-M2.7 this lineage
Youssofal/MiniMax-M2.7-abliterated-BF16↓ 63 8 forks
mradermacher/MiniMax-M2.7-abliterated-BF16-i1-GGUF↓ 719
mradermacher/MiniMax-M2.7-abliterated-BF16-GGUF↓ 272
Youssofal/MiniMax-M2.7-Abliterated-Heretic-MLX-4bit↓ 193
Youssofal/MiniMax-M2.7-Abliterated-Heretic-MLX-3bit↓ 163
lhca521/MiniMax-M2.7-abliterated-heretic-ara-AWQ↓ 60
LittleNicky55/MiniMax-M2.7-abliterated-Heretic-FP8↓ 48
ymsf/MiniMax-M2.7-abliterated-BF16-GGUF↓ 46
dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8↓ 19

Metadata

License
other
Tags
transformers safetensors minimax_m2 text-generation minimax moe abliterated heretic ara bf16 custom_code conversational

Related

Other by Youssofal
Qwen3.6-35B-A3B-Abliterated-Heretic-GGUF 7K Qwen3.6-27B-Abliterated-Heretic-Uncensored-GGUF 4K Qwen3.6-35B-A3B-Abliterated-Heretic-MLX-4bit 2K MiniMax-M2.7-Abliterated-Heretic-GGUF 975 Qwen3.6-27B-Abliterated-Heretic-Uncensored-MLX-4bit 774
Same base model
Youssofal/MiniMax-M2.7-Abliterated-Heretic-GGUF 975 Yaovi78/MiniMax-M2.7-Abliterated-Heretic-GGUF 114 lhca521/MiniMax-M2.7-Abliterated-Heretic-GGUF 96 snoweddy/MiniMax-M2.7-Abliterated-Heretic-Q8_0-GGUF 11
Total size
426 GB
Files
106
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-04-12 22:23
Source huggingface.co/Youssofal/MiniMax-M2.7-abliterated-BF16 ↗ Base model MiniMaxAI/MiniMax-M2.7 ↗

Files by quantization

Auxiliary files 106 files 426 GB
model-00001-of-00092.safetensors 4.66 GB 1e2e84d9 download
model-00084-of-00092.safetensors 4.65 GB df470742 download
model-00037-of-00092.safetensors 4.65 GB 1ba16ae3 download
model-00040-of-00092.safetensors 4.65 GB f31adf4e download
model-00062-of-00092.safetensors 4.65 GB 9cc28f7a download
model-00087-of-00092.safetensors 4.65 GB 8bf67424 download
model-00018-of-00092.safetensors 4.65 GB 204a71af download
model-00065-of-00092.safetensors 4.65 GB ebd95d85 download
model-00043-of-00092.safetensors 4.65 GB af8c7621 download
model-00090-of-00092.safetensors 4.65 GB 5440644d download
model-00021-of-00092.safetensors 4.65 GB 4b7570b5 download
model-00068-of-00092.safetensors 4.65 GB 0221da62 download
model-00046-of-00092.safetensors 4.65 GB c6f7561d download
model-00024-of-00092.safetensors 4.65 GB d782ffac download
model-00071-of-00092.safetensors 4.65 GB 731743bd download
model-00049-of-00092.safetensors 4.65 GB 96c1a2c2 download
model-00027-of-00092.safetensors 4.65 GB bd669329 download
model-00074-of-00092.safetensors 4.65 GB 43c3888d download
model-00052-of-00092.safetensors 4.65 GB e21c52b2 download
model-00077-of-00092.safetensors 4.65 GB a22a68ed download
model-00030-of-00092.safetensors 4.65 GB 6838915b download
model-00055-of-00092.safetensors 4.65 GB 8384316e download
model-00033-of-00092.safetensors 4.65 GB 1cefff92 download
model-00080-of-00092.safetensors 4.65 GB 29e2e266 download
model-00058-of-00092.safetensors 4.65 GB 35a1a577 download
model-00036-of-00092.safetensors 4.65 GB dcd13bde download
model-00083-of-00092.safetensors 4.65 GB 2570bf37 download
model-00061-of-00092.safetensors 4.65 GB c8e76b2f download
model-00039-of-00092.safetensors 4.65 GB bb2e7b97 download
model-00086-of-00092.safetensors 4.65 GB c1cf0824 download
model-00017-of-00092.safetensors 4.65 GB 1305d55d download
model-00020-of-00092.safetensors 4.65 GB d00ad1a2 download
model-00023-of-00092.safetensors 4.65 GB 18c50dbb download
model-00026-of-00092.safetensors 4.65 GB 15f6dfc4 download
model-00042-of-00092.safetensors 4.65 GB b702d740 download
model-00045-of-00092.safetensors 4.65 GB 7d23786a download
model-00048-of-00092.safetensors 4.65 GB c3d72ab4 download
model-00064-of-00092.safetensors 4.65 GB d4c9b760 download
model-00067-of-00092.safetensors 4.65 GB 5478dea3 download
model-00070-of-00092.safetensors 4.65 GB 53065993 download
model-00073-of-00092.safetensors 4.65 GB 8ced1bc7 download
model-00089-of-00092.safetensors 4.65 GB d042be03 download
model-00029-of-00092.safetensors 4.65 GB 230cddf0 download
model-00032-of-00092.safetensors 4.65 GB 3763f222 download
model-00035-of-00092.safetensors 4.65 GB 415cf671 download
model-00051-of-00092.safetensors 4.65 GB e7ad80af download
model-00054-of-00092.safetensors 4.65 GB 03594365 download
model-00057-of-00092.safetensors 4.65 GB d9599e88 download
model-00076-of-00092.safetensors 4.65 GB bf685d8b download
model-00079-of-00092.safetensors 4.65 GB 6d35071a download
model-00082-of-00092.safetensors 4.65 GB 0b74e511 download
model-00060-of-00092.safetensors 4.65 GB e44cd3fb download
model-00015-of-00092.safetensors 4.65 GB 3ae9409b download
model-00002-of-00092.safetensors 4.65 GB 510b3a3e download
model-00005-of-00092.safetensors 4.65 GB e02cf55c download
model-00008-of-00092.safetensors 4.65 GB 8d6e2e79 download
model-00011-of-00092.safetensors 4.65 GB dc09c218 download
model-00014-of-00092.safetensors 4.65 GB dead1619 download
model-00004-of-00092.safetensors 4.65 GB e840790b download
model-00007-of-00092.safetensors 4.65 GB 3357da6b download
model-00010-of-00092.safetensors 4.65 GB a287f036 download
model-00013-of-00092.safetensors 4.65 GB 30723ce8 download
model-00059-of-00092.safetensors 4.65 GB 11479330 download
model-00034-of-00092.safetensors 4.65 GB 65acff6e download
model-00081-of-00092.safetensors 4.65 GB a02f58c7 download
model-00056-of-00092.safetensors 4.65 GB 86d0bc06 download
model-00031-of-00092.safetensors 4.65 GB 9feeb12a download
model-00078-of-00092.safetensors 4.65 GB 7de62277 download
model-00053-of-00092.safetensors 4.65 GB d35d4800 download
model-00028-of-00092.safetensors 4.65 GB b677ae41 download
model-00075-of-00092.safetensors 4.65 GB c14623c7 download
model-00050-of-00092.safetensors 4.65 GB 0a9aae36 download
model-00025-of-00092.safetensors 4.65 GB aad6ac12 download
model-00072-of-00092.safetensors 4.65 GB a4ce0773 download
model-00047-of-00092.safetensors 4.65 GB 41f0234d download
model-00022-of-00092.safetensors 4.65 GB 343ad0c4 download
model-00069-of-00092.safetensors 4.65 GB 234b4001 download
model-00044-of-00092.safetensors 4.65 GB 353df61a download
model-00091-of-00092.safetensors 4.65 GB 09cca311 download
model-00019-of-00092.safetensors 4.65 GB 43428f40 download
model-00066-of-00092.safetensors 4.65 GB 9f1fa90c download
model-00041-of-00092.safetensors 4.65 GB ec83cd3a download
model-00088-of-00092.safetensors 4.65 GB 077e6648 download
model-00016-of-00092.safetensors 4.65 GB 99148ef6 download
model-00063-of-00092.safetensors 4.65 GB f9864e5c download
model-00038-of-00092.safetensors 4.65 GB 3c027b53 download
model-00085-of-00092.safetensors 4.65 GB 280f21fc download
model-00009-of-00092.safetensors 4.65 GB aace8d5e download
model-00006-of-00092.safetensors 4.65 GB b6d2667c download
model-00003-of-00092.safetensors 4.65 GB ddb46fa1 download
model-00012-of-00092.safetensors 4.63 GB c7ba6a49 download
model-00092-of-00092.safetensors 2.61 GB 0cac9035 download
tokenizer.json 14.8 MB e106d587 download
model.safetensors.index.json 4.47 MB 3e83f5e2 download
vocab.json 3.72 MB 3394578d download
merges.txt 2.30 MB a4449e69 download
modeling_minimax_m2.py 31.5 KB b8ba2586 download
tokenizer_config.json 11.0 KB 62024351 download
configuration_minimax_m2.py 9.96 KB fa618e64 download
chat_template.jinja 6.37 KB a09ec0dd download
config.json 1.63 KB 604ef01f download
special_tokens_map.json 1.60 KB 86de0560 download
.gitattributes 1.53 KB 52373fe2 download
added_tokens.json 1.43 KB e5a619a7 download
README.md 753 B f4290a46 download
generation_config.json 144 B f7ce0e8e download

README current version from Hugging Face

library_name: transformers
license: other
base_model: MiniMaxAI/MiniMax-M2.7
tags:

  • minimax
  • minimax_m2
  • moe
  • abliterated
  • heretic
  • ara
  • bf16
  • custom_code
  • text-generation

MiniMax-M2.7-abliterated-BF16

A BF16 abliterated MiniMax-M2.7 checkpoint produced with Heretic ARA.

Abliteration parameters:

  • start_layer_index: 30
  • end_layer_index: 51
  • preserve_good_behavior_weight: 0.4512
  • steer_bad_behavior_weight: 0.0037
  • overcorrect_relative_weight: 0.8804
  • neighbor_count: 14

Notes:

  • Base model: MiniMaxAI/MiniMax-M2.7
  • BF16 export repaired after save by replacing 8 non-finite sparse-MoE expert down-projection tensors from the base model
  • GGUF quants are published separately in Youssofal/MiniMax-M2.7-abliterated-GGUF

README history 1 version

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-04-12Add files using upload-large-folder tool474cc70753 B
    Loading...

Discussions 3 threads

  1. 2026-05-01NVFP4open1 💬#3
    Loading...
  2. 2026-04-23Abliteration Text Issuesopen5 💬#2
    Loading...
  3. 2026-04-14KL Divergenceopen3 💬#1
    Loading...

Continue on Hugging Face →

Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration