Skip to content

Commit 3f4e446

Browse files
gallery: add Qwopus3.6 27B Fusion variants (#11257)
Add Q4_K_M and Q8_0 llama.cpp entries for the newly released Qwopus3.6-27B Fusion reasoning and coding merge, with MTP enabled. Assisted-by: Codex:gpt-5 [Hugging Face API] Co-authored-by: localai-org-maint-bot <306269227+localai-org-maint-bot@users.noreply.github.com>
1 parent e6b235b commit 3f4e446

1 file changed

Lines changed: 68 additions & 0 deletions

File tree

gallery/index.yaml

Lines changed: 68 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -996,6 +996,74 @@
996996
- filename: llama-cpp/mmproj/Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-MTP-Q4_K_M/mmproj-F32.gguf
997997
sha256: fdc443e974cad1f61c45af1cfd5580855855ddce0d6c14cc500a5714c486ac1d
998998
uri: https://huggingface.co/DavidAU/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-NEO-MAX-MTP-GGUF/resolve/main/mmproj-F32.gguf
999+
- &qwopus3-6-27b-fusion
1000+
name: "qwopus3.6-27b-fusion"
1001+
variants:
1002+
- model: qwopus3.6-27b-fusion-q8
1003+
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
1004+
urls:
1005+
- https://huggingface.co/Qwen/Qwen3.6-27B
1006+
- https://huggingface.co/KyleHessling1/Qwopus3.6-27B-Fusion-GGUF
1007+
description: |
1008+
Qwopus3.6-27B Fusion is an experimental 27B Qwen3.6 merge that combines the
1009+
Qwopus reasoning and coding fine-tunes. It targets agentic coding and
1010+
general reasoning, supports the Qwen3.6 262K context window, and retains
1011+
the base model's MTP head. This default entry uses the 16.8 GB Q4_K_M GGUF.
1012+
license: qwen
1013+
tags:
1014+
- llm
1015+
- gguf
1016+
- reasoning
1017+
- thinking
1018+
- mtp
1019+
overrides:
1020+
backend: llama-cpp
1021+
function:
1022+
automatic_tool_parsing_fallback: true
1023+
grammar:
1024+
disable: true
1025+
known_usecases:
1026+
- chat
1027+
options:
1028+
- use_jinja:true
1029+
- spec_type:draft-mtp
1030+
- spec_n_max:6
1031+
- spec_p_min:0.75
1032+
parameters:
1033+
model: llama-cpp/models/Qwopus3.6-27B-Fusion-Q4_K_M/Qwopus3.6-27B-Fusion-Q4_K_M.gguf
1034+
template:
1035+
use_tokenizer_template: true
1036+
files:
1037+
- filename: llama-cpp/models/Qwopus3.6-27B-Fusion-Q4_K_M/Qwopus3.6-27B-Fusion-Q4_K_M.gguf
1038+
sha256: 6d4e6e9ffba84ffe5166969303d4918f3a31682d75862416827412cee8f20d5d
1039+
uri: huggingface://KyleHessling1/Qwopus3.6-27B-Fusion-GGUF/Qwopus3.6-27B-Fusion-Q4_K_M.gguf
1040+
- !!merge <<: *qwopus3-6-27b-fusion
1041+
name: "qwopus3.6-27b-fusion-q8"
1042+
variants: null
1043+
description: |
1044+
Qwopus3.6-27B Fusion in the higher-fidelity 29.0 GB Q8_0 GGUF format.
1045+
It retains the base model's MTP head for speculative decoding.
1046+
overrides:
1047+
backend: llama-cpp
1048+
function:
1049+
automatic_tool_parsing_fallback: true
1050+
grammar:
1051+
disable: true
1052+
known_usecases:
1053+
- chat
1054+
options:
1055+
- use_jinja:true
1056+
- spec_type:draft-mtp
1057+
- spec_n_max:6
1058+
- spec_p_min:0.75
1059+
parameters:
1060+
model: llama-cpp/models/Qwopus3.6-27B-Fusion-Q8_0/Qwopus3.6-27B-Fusion-Q8_0.gguf
1061+
template:
1062+
use_tokenizer_template: true
1063+
files:
1064+
- filename: llama-cpp/models/Qwopus3.6-27B-Fusion-Q8_0/Qwopus3.6-27B-Fusion-Q8_0.gguf
1065+
sha256: 5594e1776b75beedf4a54b933bba386dc83a0883417e3bcd9ef53fdfd120d5b6
1066+
uri: huggingface://KyleHessling1/Qwopus3.6-27B-Fusion-GGUF/Qwopus3.6-27B-Fusion-Q8_0.gguf
9991067
- name: "minicpm5-1b-claude-opus-fable5-v2-thinking"
10001068
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
10011069
urls:

0 commit comments

Comments
 (0)