Downloads · 30 days
18
42% of all-time downloads
Ternary-Bonsai-27B-unpacked-heretic
dfsfdsfd/Ternary-Bonsai-27B-unpacked-heretic
Ternary-Bonsai-27B-unpacked-heretic is a machine learning model from dfsfdsfd. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
[!TIP] This model is reproducible! See the README in the reproduce directory for more information.
All-time downloads
43
Public
Parameters
27.4B
54.8 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors54.7 GB · 100%
At a glance
- License
- apache-2.0
- Model type
- qwen3_5
- Access
- Public
- Created
- Aug 12, 2026
- Updated
- Aug 12, 2026
- SHA
- 5b153cd2
Model card
From the Hugging Face model README
This is a decensored version of prism-ml/Ternary-Bonsai-27B-unpacked, made using Heretic v1.4.0
[!TIP] This model is reproducible!
See the README in the
reproducedirectory for more information.
Abliteration parameters
| Parameter | Value |
|---|---|
| direction_index | 39.01 |
| attn.o_proj.max_weight | 1.28 |
| attn.o_proj.max_weight_position | 38.28 |
| attn.o_proj.min_weight | 0.29 |
| attn.o_proj.min_weight_distance | 33.83 |
| mlp.down_proj.max_weight | 1.47 |
| mlp.down_proj.max_weight_position | 38.52 |
| mlp.down_proj.min_weight | 1.00 |
| mlp.down_proj.min_weight_distance | 30.86 |
Performance
| Metric | This model | Original model (prism-ml/Ternary-Bonsai-27B-unpacked) |
|---|---|---|
| KL divergence | 0.0009 | 0 (by definition) |
| Refusals | 9/100 | 78/100 |
Ternary Bonsai 27B — Unpacked FP16 Safetensors
FP16 safetensors (HuggingFace format) of the Ternary Bonsai 27B model. This repo exists for users who want to run Ternary Bonsai with stock HuggingFace tooling or frameworks that don't yet support the packed ternary format. The 2-bit hybrid-attention kernels are currently in our forks of MLX, mlx-swift, and llama.cpp — once they land upstream, this unpacked version will no longer be needed.
We strongly recommend using the natively packed models instead. The packed format is where all the benefits of Bonsai come from — a 7.2 GB deployed footprint (down from 54 GB), 95% of FP16 intelligence retained, and interactive decoding on everyday laptops (26 tok/s on an M5 Pro). This unpacked FP16 version is full-size and does not provide any of those advantages.
For the optimized ternary release models (recommended):
- Ternary-Bonsai-27B-mlx-2bit — Ternary MLX for Apple Silicon
- Ternary GGUF (Q2_0_g128) for llama.cpp (CUDA, Metal, CPU)
For the phone-class variant:
- Bonsai-27B-mlx-1bit — 1-bit Bonsai 27B (~3.9 GB), fits an iPhone 17 Pro Max
- Type
- qwen3_5
- License
- apache-2.0
- Created
- Aug 12, 2026
- Updated
- Aug 12, 2026