MoonMath-ai

GLM-5.3-Uncensored-FP8

README

License: glm-5.3

Model description

Abliteration reduces a model's learned refusal behavior by intervening in its internal representations, without retraining. Our recipe is surgical: it modifies a small, targeted set of modules rather than broadly suppressing refusal directions:

attention output projections (self_attn.o_proj) and shared-expert projections (mlp.shared_experts.down_proj)
29 modules across 21 layers (layers 30–62 attention, layers 34–44 shared experts)

This targeted approach keeps general capabilities intact — the modification touches a minimal part of the network instead of broad refusal-direction suppression.

Refusal benchmark

We ran the model against a cybersecurity-focused refusal prompt set (attributed to Meta Purple Llama's cybersecurity benchmark). Each response was classified as addressing the request or deflecting it:

Table with columns: Model, Answered, Deflected, Refused, Answer rate
Model	Answered	Deflected	Refused	Answer rate
MoonMath	432	88	0	83.1%
Public abliterated GLM-5.3	350	170	0	67.3%

MoonMath delivered more responses and deflected fewer requests than the public abliterated model. Neither model refused outright.

Quality benchmarks

Both models ran on the same stack (sglang, identical server flags, 8× AMD MI325X). Scoring used stock lm-evaluation-harness 0.4.13 tasks (ifeval, mmlu_pro, aime25) with no custom scorers. Head-to-head against a popular public abliterated GLM-5.3:

Table with columns: Benchmark, Items, MoonMath, Public abliterated, Δ
Benchmark	Items	MoonMath	Public abliterated	Δ
IFEval (prompt-level strict)	100 prompts	91.0	39.0	+52.0
MMLU-Pro (5-shot CoT)	200 questions	89.0	84.0	+5.0
AIME 2025	30 problems × 2 samples	73.3	63.3	+10.0

The public abliterated model also exhibited severe looping degeneration (67 confirmed degenerate generations, 53% of IFEval responses hitting the 32k-token cap); MoonMath showed 0 in the same run. Full details are in the blog post.

Usage

The checkpoint follows the GLM-5.3 FP8 layout and works with transformers and vLLM / SGLang just like the base model:

python

from transformers import AutoModelForCausalLM, AutoTokenizer model = AutoModelForCausalLM.from_pretrained(    "MoonMath-ai/GLM-5.3-Uncensored-FP8", device_map="auto", trust_remote_code=True)tokenizer = AutoTokenizer.from_pretrained("MoonMath-ai/GLM-5.3-Uncensored-FP8")

Intended uses and limitations

Intended: cybersecurity research — exploit analysis, authentication-flow review, abuse-of-service analysis, and general security workflows where refusals block legitimate work.
Limitations: abliteration trades safety calibration for helpfulness. The model may answer requests that the base model would decline. It is not safety-evaluated; do not use it as a general-purpose assistant in sensitive deployments without your own evaluation.

Use it on Zro

These weights are served on Zro, MoonMath's private inference platform, as the model Dolly 2 Security (dolly2-security). No GPU setup, and no prompts retained after processing.

With the CLI (open source, MIT):

bash

npm install -g @moonmath-ai/zrozro loginzro launch claude --model dolly2-security

zro launch also works with codex, opencode, hermes, openclaw, pi and many more harnesses.

With the API (OpenAI-compatible):

bash

curl https://zro.moonmath.ai/v1/chat/completions \  -H "Authorization: Bearer $ZRO_API_KEY" \  -H "Content-Type: application/json" \  -d '{"model":"dolly2-security","messages":[{"role":"user","content":"Reply with OK"}]}'

Model ID: dolly2-security (alias zro/dolly2-security). The lighter dolly1-security is also available.
1M-token context, up to 131K output. $2.10 p er 1 M in p u tt o k e n s,$ 6.30 per 1M output, $0.45 per 1M cache-read tokens. Prompt caching is on by default.
Inference region per API key: Europe, the United States, or both. Requests are routed through the EU.
Zero request retention, and no training on requests.
Plans are monthly spend budgets, not token caps: Pro ( $20/ m o n t h,$ 40 usage budget) and Max ( $60/ m o n t h,$ 120 usage budget). A one-time $5 start needs no subscription. Team is coming soon.

Links: pricing · CLI docs · API reference · privacy · sign up · Dolly and me

License

GLM-5.3 is released by Z.ai under the GLM-5.3 model license (see LICENSE in the base repo). This checkpoint inherits the base model's license and terms.

Citation

bibtex

@misc{moonmath2026dolly2,  title  = {Dolly and me: Building cybersecurity models at Zro},  author = {MoonMath.ai},  year   = {2026},  url    = {https://zro.moonmath.ai/posts/dolly-and-me}}

Acknowledgements

Z.ai for the GLM-5.3 base model

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

MoonMath-ai

Model Tree

Base

zai-org/GLM-5.3

Quantized

this model

Input Modalities

Text

Output Modalities

Text

Supported Functionality

Dedicated Endpoints

Explore FriendliAI today

Get started Talk to an engineer

README

License: glm-5.3

Model description

attention output projections (self_attn.o_proj) and shared-expert projections (mlp.shared_experts.down_proj)
29 modules across 21 layers (layers 30–62 attention, layers 34–44 shared experts)

This targeted approach keeps general capabilities intact — the modification touches a minimal part of the network instead of broad refusal-direction suppression.

Refusal benchmark

Table with columns: Model, Answered, Deflected, Refused, Answer rate
Model	Answered	Deflected	Refused	Answer rate
MoonMath	432	88	0	83.1%
Public abliterated GLM-5.3	350	170	0	67.3%

MoonMath delivered more responses and deflected fewer requests than the public abliterated model. Neither model refused outright.

Quality benchmarks

Table with columns: Benchmark, Items, MoonMath, Public abliterated, Δ
Benchmark	Items	MoonMath	Public abliterated	Δ
IFEval (prompt-level strict)	100 prompts	91.0	39.0	+52.0
MMLU-Pro (5-shot CoT)	200 questions	89.0	84.0	+5.0
AIME 2025	30 problems × 2 samples	73.3	63.3	+10.0

Usage

The checkpoint follows the GLM-5.3 FP8 layout and works with transformers and vLLM / SGLang just like the base model:

python

from transformers import AutoModelForCausalLM, AutoTokenizer model = AutoModelForCausalLM.from_pretrained(    "MoonMath-ai/GLM-5.3-Uncensored-FP8", device_map="auto", trust_remote_code=True)tokenizer = AutoTokenizer.from_pretrained("MoonMath-ai/GLM-5.3-Uncensored-FP8")

from transformers import AutoModelForCausalLM, AutoTokenizer model = AutoModelForCausalLM.from_pretrained(    "MoonMath-ai/GLM-5.3-Uncensored-FP8", device_map="auto", trust_remote_code=True)tokenizer = AutoTokenizer.from_pretrained("MoonMath-ai/GLM-5.3-Uncensored-FP8")

Intended uses and limitations

Intended: cybersecurity research — exploit analysis, authentication-flow review, abuse-of-service analysis, and general security workflows where refusals block legitimate work.
Limitations: abliteration trades safety calibration for helpfulness. The model may answer requests that the base model would decline. It is not safety-evaluated; do not use it as a general-purpose assistant in sensitive deployments without your own evaluation.

Use it on Zro

These weights are served on Zro, MoonMath's private inference platform, as the model Dolly 2 Security (dolly2-security). No GPU setup, and no prompts retained after processing.

With the CLI (open source, MIT):

bash

npm install -g @moonmath-ai/zrozro loginzro launch claude --model dolly2-security

npm install -g @moonmath-ai/zrozro loginzro launch claude --model dolly2-security

zro launch also works with codex, opencode, hermes, openclaw, pi and many more harnesses.

With the API (OpenAI-compatible):

bash

curl https://zro.moonmath.ai/v1/chat/completions \  -H "Authorization: Bearer $ZRO_API_KEY" \  -H "Content-Type: application/json" \  -d '{"model":"dolly2-security","messages":[{"role":"user","content":"Reply with OK"}]}'

curl https://zro.moonmath.ai/v1/chat/completions \  -H "Authorization: Bearer $ZRO_API_KEY" \  -H "Content-Type: application/json" \  -d '{"model":"dolly2-security","messages":[{"role":"user","content":"Reply with OK"}]}'

Model ID: dolly2-security (alias zro/dolly2-security). The lighter dolly1-security is also available.
1M-token context, up to 131K output. $2.10 p er 1 M in p u tt o k e n s,$ 6.30 per 1M output, $0.45 per 1M cache-read tokens. Prompt caching is on by default.
Inference region per API key: Europe, the United States, or both. Requests are routed through the EU.
Zero request retention, and no training on requests.
Plans are monthly spend budgets, not token caps: Pro ( $20/ m o n t h,$ 40 usage budget) and Max ( $60/ m o n t h,$ 120 usage budget). A one-time $5 start needs no subscription. Team is coming soon.

Links: pricing · CLI docs · API reference · privacy · sign up · Dolly and me

License

GLM-5.3 is released by Z.ai under the GLM-5.3 model license (see LICENSE in the base repo). This checkpoint inherits the base model's license and terms.

Citation

bibtex

@misc{moonmath2026dolly2,  title  = {Dolly and me: Building cybersecurity models at Zro},  author = {MoonMath.ai},  year   = {2026},  url    = {https://zro.moonmath.ai/posts/dolly-and-me}}

@misc{moonmath2026dolly2,  title  = {Dolly and me: Building cybersecurity models at Zro},  author = {MoonMath.ai},  year   = {2026},  url    = {https://zro.moonmath.ai/posts/dolly-and-me}}

Acknowledgements

Z.ai for the GLM-5.3 base model