base_model:
- huihui-ai/Huihui-Qwen3.8-Flash-Next-abliterated base_model_relation: quantized license: other license_name: qwen-community-1.0 library_name: gguf pipeline_tag: image-text-to-text tags:
- gguf
- qwen3.8-flash-next
- qwen4exp
- abliterated
- uncensored
- experimental
- agentionai
Qwen3.8-Flash-Next Gyro, abliterated by huihui-ai — GGUF (experimental)
The abliterated Qwen3.8-Flash-Next on a single GPU, in two sizes.
This is huihui-ai/Huihui-Qwen3.8-Flash-Next-abliterated encoded the same way as our Gyro files. Abliteration is huihui-ai's work: it removes the refusal direction from the model's weights. We only quantized it.
| File | For | GPU memory (64k context) | KLD vs the original model |
|---|---|---|---|
Qwen3.8-Flash-Next-Gyro-S-abliterated-TQ1_0.gguf | 32 GB cards | 28.8 GiB | 0.435 (unchanged from Gyro-S) |
Qwen3.8-Flash-Next-Gyro-M-abliterated-TQ2_0.gguf | 48 GB: 2×24 GB cards, 64 GB unified-memory machines | ~37 GiB | 0.310 (Gyro-M: 0.306) |
Which setup fits your hardware, NVIDIA and AMD instructions, context limits, vision and speculative decoding are all on the main Gyro card and apply unchanged here; only the file names differ.
Same architecture, size, speed and runtime as Gyro-S and Gyro-M. Abliteration changes 144 of the model's 1,224
weight tensors (the ones that write into the residual stream: attention and linear-attention outputs, and the down
projections of the shared and routed experts); everything else is byte-identical to the Gyro files. The TQ1_0 /
TQ2_0 suffix is a size class for the Hub's file browser, not the format: these files use our own rotor formats.
Usage warning
Abliteration removes most of the model's refusals. Like huihui-ai's original, this model can produce harmful, offensive or inappropriate content, and it is not suitable for public-facing or unsupervised use. You are responsible for how you use it and for complying with the law where you are. See huihui-ai's model card for their full warnings.
Quick start (Docker)
The rotor code needs our llama.cpp build, agentionai/llama.cpp (Vulkan). The container has it ready:
hf download agentionai/Qwen3.8-Flash-Next-Gyro-abliterated-Huihui-GGUF Qwen3.8-Flash-Next-Gyro-S-abliterated-TQ1_0.gguf \
mtp-Qwen3.8-Flash-Next-draft.gguf mmproj-F16.gguf --local-dir ~/models
docker run --rm -it --device /dev/dri --group-add "$(getent group render | cut -d: -f3)" \
-v ~/models:/models -p 8080:8080 ghcr.io/agentionai/agention-llama:server \
-m /models/Qwen3.8-Flash-Next-Gyro-S-abliterated-TQ1_0.gguf -ngl 999 -fa on --jinja --ngram-on-disk \
-c 65536 -ctk q8_0 -ctv q8_0 --temp 1.0 --top-p 0.95 --top-k 20 --min-p 0 \
--mmproj /models/mmproj-F16.gguf
Or with -hf agentionai/Qwen3.8-Flash-Next-Gyro-abliterated-Huihui-GGUF:Gyro-S in a native build, which also fetches the
vision projector and, with --spec-type draft-mtp, the MTP draft. NVIDIA, vision on/off, speculative decoding and
memory per context length: exactly as on the Gyro card.
The MTP draft was made from the original model; it still drafts well for the abliterated weights (81% of drafted
tokens accepted on a code prompt in a first check).
Quality
KL divergence and top-1 agreement on our held-out corpus (2026 technical writing and code, -c 2048, 60 chunks),
measured against the original (not abliterated) Qwen3.8-Flash-Next at Q8_0. These numbers therefore include the
abliteration itself, not only the quantization.
On this corpus abliteration costs nothing measurable: abliterated Gyro-S matches Gyro-S to within the error bars (0.4351 vs 0.4352 KL divergence, ±0.003). That is expected for technical text and code, which never touch the refusal direction abliteration removes. It says nothing about refusals themselves; for that, see huihui-ai's card.
| File | GPU weights | KLD vs original | top-1 vs original |
|---|---|---|---|
| Gyro-S (original model) | 27.15 GiB | 0.435 | 74.1 % |
| abliterated Gyro-S | 27.15 GiB | 0.435 | 74.1 % |
| Gyro-M (original model) | 34.35 GiB | 0.306 | 78.1 % |
| abliterated Gyro-M | 34.35 GiB | 0.310 | 78.1 % |
Throughput
Same as the Gyro files, which have identical tensor types and sizes: Gyro-S on one AMD Radeon AI PRO R9700 32 GB,
prefill 1,246 / decode 57.8 tok/s; on AMD Strix Halo (Radeon 8060S), 250 / 31.7 tok/s; Gyro-M on Strix Halo, 249 / 26.4
(llama-bench, batch 1, no drafting).
Status
Experimental, like the Gyro files: the format and kernels may change. Vulkan only; tested on AMD Strix Halo. The weights come from our own encoder and calibration, which we are not publishing at this stage. The calibration was collected on the original model and reused here.
Credits and license
Base model: Qwen/Qwen3.8-Flash-Next (Qwen community license 1.0, see LICENSE). Abliteration: huihui-ai. Quantization: Agention.
Support AgentionAI
These quants are released freely. If they save you VRAM or make Qwen more useful, you can buy me a coffee or some GPU time and sponsor continued quantization and benchmarking on GitHub. AgentionAI is a one-person team and can use your help.