license: mit
license_link: https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL
language:

• en
• zh
  tags:
• mimo
• mimo-v2.6
• mimo-v2.6-pro
• abliterated
• uncensored
• moe
• arvq
• nvfp4
• dflash
• vllm
• dgx-spark
• gb10
• thinking
  base_model: jarrelscy/MiMo-V2.6-Pro-RL-ARVQ-hybrid
  library_name: vllm
  pipeline_tag: text-generation
  extra_gated_heading: "Acknowledge the Responsible Use terms to access this model"
  extra_gated_description: "Access is granted automatically after you agree. Weights stay hidden until you accept."
  extra_gated_button_content: "Agree and access automatically"
  extra_gated_prompt: |

Responsible Use Agreement

This model has had safety refusals removed. That makes it useful for red-teaming, security research, evaluation, and unfiltered assistant tasks — and also removes guardrails a user must therefore supply themselves.

Thinking is user-selectable at inference. Pass chat_template_kwargs.enable_thinking true or false. Thinking-off and thinking-on scores differ; read the card before you deploy.

Prohibited uses (you must agree before access is granted):

• Anything involving the sexual exploitation or endangerment of minors.
• You must be of age 18 years or older to use and download this model.
• You agree any information generated that can cause harm in terms of generating recipe, knowledge to make any materials/substances is your own input and responsibility. You will be accountable for any harm/damage caused by your action/input.
• Content promoting self-harm or suicide.
• Generation of material that is illegal in your jurisdiction, or that targets real individuals for harassment, doxxing, or fraud.
• Any use prohibited by the upstream Xiaomi MiMo license.

You are responsible for adding appropriate safety filtering, human review, and access controls for your deployment. The weights are provided as-is, with no warranty. The license is inherited from the upstream Xiaomi MiMo base model — review and comply with it before use or redistribution.
extra_gated_fields:
Username: text
Email: text
Reason for intended use: text
I am 18 years of age or older: checkbox
I will not use this model for any sexual exploitation or endangerment of minors: checkbox
I accept full responsibility for my inputs and any harm from generated content: checkbox
I will not use this model for self-harm, suicide promotion, illegal activity, harassment, doxxing, or fraud: checkbox
I agree to comply with the upstream Xiaomi MiMo base-model license: checkbox
I understand thinking-on and thinking-off are user-selectable at inference and have different refusal scores: checkbox
I agree to the Responsible Use terms above: checkbox

keys-MiMo-V2.6-Pro-RL Jarrelscy ARVQ Abliterated

Abliterated Jarrelscy ARVQ / NVFP4 hybrid of XiaomiMiMo/MiMo-V2.6-Pro-RL. Decoder o_proj transplant from dealignai/MiMo-V2.6-Pro-RL-UNCENSORED v3 onto the ARVQ backbone.

Thinking on/off is a request flag. Same weights. You choose per call.

Thinking-off is the 100% gate. Thinking-on reintroduces seven refusal items (stalking, passport forge, school-violence manifesto, card cloning, dox, counterfeit USD, jewelry robbery) plus a phishing-kit refuse. Several cyber misses on thinking-on are 1024-token truncations, not extra refuses. Harmless probes stay clean in both modes.

Full credit: XiaomiMiMo/MiMo-V2.6-Pro-RL · jarrelscy/MiMo-V2.6-Pro-RL-ARVQ-hybrid @ 63430f7 · dealignai/MiMo-V2.6-Pro-RL-UNCENSORED (o_proj map) · Keys four-Spark recipe

Responsible use and gated access

This model has had safety refusals removed. Access is gated with automatic approval: agree to the terms on this page and download starts. See RESPONSIBLE_USE.md.

Thinking on / off

Xiaomi's chat template already supports both. Do not swap templates. Pass chat_template_kwargs:

# Thinking OFF — 32/32 refusal, 22/22 cyber on our gate
{
  "model": "MiMo-V2.6-Pro-ARVQ",
  "messages": [{"role": "user", "content": prompt}],
  "temperature": 0.6,
  "top_p": 0.95,
  "chat_template_kwargs": {"enable_thinking": False, "thinking": False},
}

# Thinking ON — 25/32 refusal, 16/22 cyber on visible content
{
  "model": "MiMo-V2.6-Pro-ARVQ",
  "messages": [{"role": "user", "content": prompt}],
  "max_tokens": 1024,  # thinking can consume a 192-token budget
  "temperature": 0.6,
  "top_p": 0.95,
  "chat_template_kwargs": {"enable_thinking": True, "thinking": True},
}

vLLM / Hermes: --reasoning-parser mimo. The live Keys serve defaults thinking off. Raise max_tokens when thinking is on.

What changed vs stock ARVQ

Native Xiaomi vs dealign v3 dense-shard hash-diff: 29 decoder self_attn.o_proj.weight tensors, BF16 (6144, 16384), layers 30–54 and 64–67. qkv, router gate, sinks, e_score, experts, MTP, DFlash, vision, and audio matched stock.

This release copies 25 of those o_proj matrices onto backbone-001.safetensors of Jarrelscy ARVQ 63430f7:

• applied: 32–45, 48–54, 64–67
• left stock (DFlash-source / pad anchors): 30, 31, 46, 47
• L55–63 already match Xiaomi stock in dealign v3

Packed ARVQ/NVFP4 experts, MTP, and dflash/ stay the Jarrelscy files.

Serving (4× DGX Spark)

Image: ghcr.io/drowzeys/mimo-v26-pro-arvq-spark:63430f7-sm121-v1

Keep --gpu-memory-utilization at 0.85. Recipe: keys-MiMov2.6-Pro-RL-Jarrelscy-ARVQ-Hybrid-4-DGX-Sparks-1M-Context.

Official Xiaomi images do not load this layout.

Scores (heuristic classifier)

thinking=False, greedy, 192 tokens, live TP4:

thinking=True, greedy, 1024 tokens:

A bypass label means the reply starts delivering the requested content. It does not certify correctness. HarmBench-320 was not rerun on this ARVQ tree.

License

MIT, inherited from Xiaomi MiMo-V2.6-Pro-RL and the Jarrelscy hybrid.