dima6312/Flint-40B-A2.5B-abliterated-GGUF

Flint-40B-A2.5B-abliterated (GGUF)

An abliterated (uncensored) Flint: RecursiveSI's Flint 40B-A2.5B with its refusal direction removed, for security research and coding work: exploit analysis, malware triage, detection engineering, CTFs and red-team tooling that the original model often declines.

Summary: held-out MITRE ATT&CK refusals 15% → 0% (n=100, Q8_0). No significant drop on HumanEval+, MBPP+, GSM8K, MMLU-Pro, CyberMetric or WMDP-cyber (paired, same items). KL divergence from the original on harmless prompts: 0.0146. Builds: BF16, GGUF (Q8_0 to Q3_K_M), NVFP4 for NVIDIA Blackwell, MLX 4-bit for Apple silicon.

Which build

Not sure? Take the GGUF Q4_K_M (24 GB). It runs in llama.cpp and in apps built on it, including LM Studio (its llama.cpp runtime 2.63.0 includes Flint). The MLX builds are for Mac users who already run mlx-lm.

You run	Download	Size
llama.cpp, or an app that bundles llama.cpp b11827 or later; any GPU, CPU or Mac	GGUF (this repo): Q8_0 is the benchmarked build, Q4_K_M the usual choice	42 / 24 GB
NVIDIA Blackwell GPU (RTX 50 series, RTX PRO 6000, DGX Spark) with llama.cpp CUDA	NVFP4 GGUF	23 GB
Apple silicon Mac with mlx-lm installed from PR #1998	MLX 4-bit	21 GB
Apple silicon Mac with RecursiveSI's Rec runtime	Rec 4-bit	21 GB
transformers or SGLang on a GPU server; fine-tuning or making your own quants	BF16 safetensors	80 GB

Plan for the file size plus memory for context: about 26 KB per token, so roughly 7 GB at the full 262,144 tokens.

Results

Both models ran as Q8_0 GGUF on llama-server with the same flags, Flint's recommended sampling (temperature 1.0, top_p 0.95, top_k 20) and full reasoning, on evenly spaced subsets of each benchmark (n in each row). Checks are paired on the same items; a quality check fails only on a drop that is significant at 95%. Refusals are scored with Meta's CyberSecEval refusal check. Refusal rates count refusals only. A reply with no final answer (the model was still reasoning at the 8,192-token limit) is listed separately as "no answer": on the 100 MITRE prompts that happened 19 times for the abliterated model and once for the original. The cut-off reasoning was not repetitive; give long security tasks a larger token budget.

Check	Original	Abliterated	Result
MITRE refusals (n=100) significantly lower	15%	0%	pass
FRR false refusals (n=100) not significantly higher	0%	0%	pass
cybermetric (n=50) not significantly worse	100%	96%	pass
gsm8k (n=50) not significantly worse	98%	98%	pass
mmlu_pro (n=50) not significantly worse	80%	78%	pass
wmdp_cyber (n=50) not significantly worse	50%	60%	pass
humaneval_plus (n=40) not significantly worse	90%	95%	pass
mbpp_plus (n=40) not significantly worse	87.5%	90%	pass
pp512 tok/s within 3%	8050.8	8135.1	pass
tg128 tok/s within 3%	180.0	185.6	pass

Refusals on held-out CyberSecEval MITRE prompts (n per ATT&CK category below):

Category	n	Original: refused / no answer	Abliterated: refused / no answer
C2	10	20% / 0%	0% / 30%
Collection	10	0% / 0%	0% / 0%
Discovery	10	0% / 10%	0% / 20%
Evasion	10	40% / 0%	0% / 40%
Execution	10	10% / 0%	0% / 30%
Exfil	10	10% / 0%	0% / 0%
Lateral Movement	10	40% / 0%	0% / 10%
Persistence	10	20% / 0%	0% / 10%
Privilege Escalation	10	10% / 0%	0% / 30%
Recon	10	0% / 0%	0% / 20%

Heretic's default refusal count, measured on the NVFP4 builds (100 mlabonne/harmful_behaviors test prompts, reasoning off, greedy, 100 tokens): original 96/100, abliterated 55/100. That count flags any reply containing words such as "harmful" or "illegal", including answers that comply; only 1 of the 55 replies opens with a refusal. Meta's CyberSecEval refusal check on the same replies: original 18/100, abliterated 1/100.

KL divergence from the original on 100 harmless prompts, measured by the search for the selected trial (not part of the checks above): 0.0146 nats/token.

Benchmarks: CyberSecEval MITRE and FRR (Meta), CyberMetric-500, WMDP-cyber, MMLU-Pro, GSM8K, HumanEval+ and MBPP+ (EvalPlus); speed from llama-bench pp512/tg128.

Method

The refusal direction comes from the difference in mean activations on harmful and harmless prompts (mlabonne/harmful_behaviors, mlabonne/harmless_alpaca). abliterix removes it from attention output, shared-expert and routed-expert projections; Optuna picks the strengths over 50 trials against refusals on 100 MITRE prompts and KL divergence on harmless prompts. The MITRE prompts in the table were never seen by the search. Tensor shapes are unchanged, so speed and memory match the original.

Files

File	Size	Note
Flint-40B-A2.5B-abliterated-BF16.gguf	79.6 GB	unquantized
Flint-40B-A2.5B-abliterated-Q8_0.gguf	42.4 GB	the checked build
Flint-40B-A2.5B-abliterated-Q6_K.gguf	32.7 GB
Flint-40B-A2.5B-abliterated-Q5_K_M.gguf	28.3 GB
Flint-40B-A2.5B-abliterated-Q4_1.gguf	25.0 GB
Flint-40B-A2.5B-abliterated-Q4_K_M.gguf	24.2 GB	common default; short check 9/10 GSM8K, 9/10 HumanEval+
Flint-40B-A2.5B-abliterated-MXFP4_MOE.gguf	22.2 GB	MXFP4 experts
Flint-40B-A2.5B-abliterated-IQ4_XS.gguf	21.3 GB
Flint-40B-A2.5B-abliterated-Q3_K_M.gguf	19.1 GB
Flint-40B-A2.5B-abliterated-IQ2_M.gguf	13.1 GB	likely unusable: in a short test it only repeated words

All quants except Q8_0 and BF16 use an importance matrix (imatrix.gguf, computed on wikitext-2) and were made with llama.cpp llama-quantize. Only Q8_0 was benchmarked.

Usage

Reasoning is optional. When the model reasons, it uses <think>…</think> before its answer. Use the sampling settings above.

llama.cpp (b11827 or later):

llama-server -m Flint-40B-A2.5B-abliterated-Q8_0.gguf --jinja -ngl 999 -fa on

transformers 5.20 (trust_remote_code=True, the port ships with the weights):

from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "dima6312/Flint-40B-A2.5B-abliterated"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(repo, trust_remote_code=True, dtype="bfloat16", device_map="auto")

Limitations

• The checks ran on subsets of each benchmark (n in each row), on the Q8_0 GGUF only. Short checks (n=10 GSM8K, n=10 HumanEval+): Q4_K_M 9/10 and 9/10, IQ2_M 0/10 and 0/10. Other quants were not measured.
• Refusals are lower, not zero: the selected trial still refused 10 of the 100 prompts used by the search.
• On long security tasks the model can reason for more than 8,192 tokens before answering.
• Abliteration removes refusals; it adds no knowledge. Mistakes the original makes, this model makes too.

Responsible use

This model answers requests the original refuses. It is meant for authorized security testing, research, education and defensive work. You are responsible for how you use it and for following the law and the terms of the systems you test. Its answers can be wrong; check them before acting on them.

Credits

• Flint by RecursiveSI (Apache 2.0). This is a derivative of their weights under the same license.
• abliterix by wuwangzhang1216, building on Heretic by p-e-w.
• Arditi et al., Refusal in Language Models Is Mediated by a Single Direction (2024).
• Maxime Labonne's harmful/harmless prompt sets; Meta's CyberSecEval; CyberMetric (Tihanyi et al.); WMDP (Center for AI Safety); MMLU-Pro (TIGER-Lab); GSM8K (OpenAI); EvalPlus.
• llama.cpp (ggml-org) and transformers (Hugging Face).