websfactory

Webs-KoReasoner-27B-v1

README

License: apache-2.0

Recipe

Base: Qwen/Qwen3.5-27B (Apache-2.0)
Donors (same base):
- NewenAI/QuettaLLMs-27B-Koreasoner-V3
- jiwon9703/Qwen3.5-KoReasoin-27B-v1
Method: DARE-TIES (density 0.5, standard 1/density rescale, weight 1.0, seed 42), TIES sign-election across the two task vectors, applied to all tensors (including MLP). fp32 math, streamed one tensor at a time so a 64GB machine never holds more than one tensor's working set.

Because mergekit cannot express the Qwen3_5 hybrid architecture (interleaved linear_attention / full_attention layers), the merge was produced with an in-house streaming merger.

Intended use

Korean knowledge & reasoning. The model thinks (often in English) inside <think> … </think> and answers in Korean.

Credits

All weights derive from the Apache-2.0 base and donors above; full credit to their authors. Merge engineering by 웹스팩토리 (Websfactory).

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

websfactory

Model Tree

Base

NewenAI/QuettaLLMs-27B-Koreasoner-V3

Base

jiwon9703/KoQweopus-3.5-27B-experimental

Base

Qwen/Qwen3.5-27B

Merged

this model

Input Modalities

TextImageVideo

Output Modalities

Text

Supported Functionality

Dedicated Endpoints

Explore FriendliAI today

Get started Talk to an engineer

README

License: apache-2.0

Recipe

Base: Qwen/Qwen3.5-27B (Apache-2.0)
Donors (same base):
- NewenAI/QuettaLLMs-27B-Koreasoner-V3
- jiwon9703/Qwen3.5-KoReasoin-27B-v1
Method: DARE-TIES (density 0.5, standard 1/density rescale, weight 1.0, seed 42), TIES sign-election across the two task vectors, applied to all tensors (including MLP). fp32 math, streamed one tensor at a time so a 64GB machine never holds more than one tensor's working set.

Because mergekit cannot express the Qwen3_5 hybrid architecture (interleaved linear_attention / full_attention layers), the merge was produced with an in-house streaming merger.

Intended use

Korean knowledge & reasoning. The model thinks (often in English) inside <think> … </think> and answers in Korean.

Credits

All weights derive from the Apache-2.0 base and donors above; full credit to their authors. Merge engineering by 웹스팩토리 (Websfactory).