Downloads · 30 days
87
13% of all-time downloads
underlotus/Qwen3.8-27B-heretic-ara-oQ3e-mtp
Qwen3.8-27B-heretic-ara-oQ3e-mtp is a image-text-to-text model from underlotus. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as apache-2.0.
Mixed-precision quant of trohrbaugh/Qwen3.8-27B-heretic-ara, produced with oQ (oMLX v0.6.2). Vision tower intact — image + video input still works. MTP head preserved. Standard MLX safetensors — compatible with oMLX,…
Downloads · 30 days
87
13% of all-time downloads
All-time downloads
685
Public
Parameters
27.8B
13.9 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors13.9 GB · 100%
How the weights are stored.
U3227.3B · 98%
From the Hugging Face model README
Mixed-precision quant of trohrbaugh/Qwen3.8-27B-heretic-ara, produced with oQ (oMLX v0.6.2). Vision tower intact — image + video input still works. MTP head preserved. Standard MLX safetensors — compatible with oMLX, mlx-lm, LM Studio, and any MLX-capable app.
Unlike uniform 3-bit quantization, oQ is a data-driven mixed-precision quantizer that calibrates per-layer sensitivity and allocates bits where they matter most. Critical layers (embeddings, LM head, the most sensitive transformer layers) are automatically promoted to higher precision, while less sensitive layers stay at 3-bit.
| Context | PP tok/s | TG tok/s | Peak Mem |
|---|---|---|---|
| 1k | 63.0 | 11.1 | 15.5 GB |
| 4k | 61.8 | 12.3 | 17.1 GB |
With Lightning MTP enabled. For reference, with MTP disabled the same quant measures ~8.0 / 7.7 TG tok/s at 1k / 4k (13.4 / 14.9 GB) — on the lower-bit quants the MTP head mostly costs memory rather than adding throughput.
| Batch | TG tok/s | Speedup |
|---|---|---|
| 1× | 11.1 | 1.00× |
| 2× | 14.5 | 1.31× |
| 4× | 24.8 | 2.23× |
Full benchmark → (1k) · 4k
The original BF16 weights require ~55 GB. This oQ3e quant runs in ~13–15 GB on Apple Silicon while keeping the vision tower and MTP head intact — the lightest option in the oQe family.
# oMLX
omlx serve --model underlotus/Qwen3.8-27B-heretic-ara-oQ3e-mtp
# mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("underlotus/Qwen3.8-27B-heretic-ara-oQ3e-mtp")
response = generate(model, tokenizer, prompt="Hello!", max_tokens=256)
print(response)
Apache 2.0, inherited from base model.