Downloads · 30 days
191
57% of all-time downloads
Qwen3.8-Flash-Next-UD-Q2_K_XL
KoenKeunen/Qwen3.8-Flash-Next-UD-Q2_K_XL
Qwen3.8-Flash-Next-UD-Q2_K_XL is a machine learning model from KoenKeunen. 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 other.
This is a merged version of the Unsloth version of Qwen 3.8 Flash Next in Q2, which can be found here: https://huggingface.co/unsloth/Qwen3.8-Flash-Next-GGUF/tree/main This merged version has two advantages for those…
All-time downloads
334
Public
Repo size
78.9 GB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.gguf78.9 GB · 100%
At a glance
- License
- other
- Access
- Public
- Created
- Sep 21, 2026
- Updated
- Sep 26, 2026
- SHA
- a746e6a1
Base models
Model card
From the Hugging Face model README
This is a merged version of the Unsloth version of Qwen 3.8 Flash Next in Q2, which can be found here: https://huggingface.co/unsloth/Qwen3.8-Flash-Next-GGUF/tree/main This merged version has two advantages for those looking to quickly run Qwen 3.8 Flash next on consumer hardware without much configuration:
- it can be installed instantly from Hugging Face in Ollama without manual local merging
- it comes without the MTP weights, and thus will also automatically run without MTP, which will only slow inference on RTX 4090 / RTX 3090 / RTX 5090 because of the lack of vram.
On a single RTX 5090, this version works at around 35 tokens/s in ollama on a system with DDR5 ram (at least 64 gb). In Unsloth Desktop, it even runs at 40 tokens/s.
- License
- other
- Created
- Sep 21, 2026
- Updated
- Sep 26, 2026