GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
akumaburn
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
akumaburn
Model Tree
akumaburn/Qwen3.8-27B-heretic
Input Modalities
Output Modalities
Supported Functionality
Source: Qwen3.8-27B-heretic (the abliterated BF16). This build changes numerics only; refusal behaviour is identical to the source (0/100 hard refusals; see the source card for the ~20–24 % chain-of-thought deflection note).
Quantization fidelity — KL divergence from the source, KL(source ‖ this),
measured served under vLLM (real INT8 dynamic activations) on WikiText-2,
top-512, 12 264 token positions:
| this build | non-abliterated sibling | |
|---|---|---|
| KL, floor-free | 0.0080 | 0.0081 |
| KL, raw (with floor) | 0.0083 | 0.0110 |
On par with the reference W8A8 of this model — the abliteration does not make the weights harder to quantize.
Speculative decoding — acceptance, HumanEval, concurrency 32, vLLM
(fp8_e4m3 KV, async scheduling, MTP num_speculative_tokens=3):
| acceptance rate | acceptance length | |
|---|---|---|
| this build | 50.1 % | 2.50 |
| non-abliterated sibling (same config) | 50.6 % | 2.52 |
Acceptance length (mean tokens accepted per verify step) sets the speculative speedup; it is unchanged from the non-abliterated model. Heretic leaves the MTP head untouched — it is bitwise identical to the source's — and the abliteration is a low-KL edit (0.088 nats) confined to 63 output projections in layers 11–63, so drafter/target agreement barely moves.
Target: 2× Ampere (SM 8.0).
sh
vllm serve akumaburn/Qwen3.8-27B-heretic-SmoothQuant-W8A8-INT8 \--tensor-parallel-size 2 \--max-model-len 262144 \--kv-cache-dtype fp8_e4m3 \--async-scheduling \--speculative-config '{"method":"mtp","num_speculative_tokens":3}'
All quantized matmul dimensions are multiples of 128 (no INT8 padding); TP=2
shards cleanly (GQA kv_heads=4 divides evenly).
Refusal behaviour has been deliberately removed; this model produces content the source declines, including dangerous, illegal, or unethical material, with no moderation. Intended for interpretability/safety research, red-teaming, and evaluation by people who understand and accept those risks. Do not deploy it where it can reach people who have not consented to unfiltered output. You are responsible for your use and for compliance with all applicable laws.
This model is provided for research purposes only, "AS IS", without warranty of any kind, express or implied. The author accepts no liability for any use of this model or any consequences arising from it. By downloading or using it, you accept sole responsibility for your use and for compliance with all applicable laws and regulations. Base model © Qwen (Apache-2.0), inherited by this derivative; abliteration © the Heretic project; quantization via llm-compressor / compressed-tensors.