mirror/LocalAI
1
0 Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2026-10-28 09:08:54 -04:00
Code Issues Packages Projects Releases Wiki Activity

feat(gallery): add Tess 4.5 27B variants

Browse Source 
Add three GGUF quantizations with a shared vision projector and pinned
artifact URLs. Document installation and the default context size.

Assisted-by: Codex:gpt-6
This commit is contained in:
localai-org-maint-bot committed 2026-10-23 10:04:46 +00:00
1 parent 7d528e90f4
commit c8df4d6fef
3 files changed
+318
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

docs/content/features/model-gallery.md
+12
View File
@@ -606,6 +606,18 @@ where:
- `bert-embeddings` is the model name in the gallery
(read its [config here](https://github.com/mudler/LocalAI/blob/master/gallery/index.yaml)).
### Tess 4.5 27B
Install `tess-4.5-27b-q4` for coding, reasoning, tool use, and image input
with llama.cpp. The gallery groups Q4_K_M, Q6_K, and Q8_0 builds as variants.
Each build includes the F16 vision projector and uses a 32,768-token context
by default. These entries do not enable MTP speculative decoding or download
the separate draft heads.
```bash
local-ai models install tess-4.5-27b-q4
```
### Model variants
Some gallery entries offer several builds of the same model: different
gallery/index.yaml
+141
View File
@@ -3,6 +3,147 @@
# written by scripts/models/cmd/modelcatalog and stays only so releases before
# the cutover keep working. CI fails when it differs from the generated copy:
# go run ./scripts/models/cmd/modelcatalog --library gallery/models/index.yaml --out gallery/index.yaml
- url: github:mudler/LocalAI/gallery/virtual.yaml@master
name: tess-4.5-27b-q4
description: |
Tess 4.5 is a 27B Qwen3.8 fine-tune for coding, reasoning, and tool use.
This Q4_K_M GGUF build includes the F16 vision projector and uses
llama.cpp with the embedded chat template. Speculative decoding is disabled.
license: apache-2.0
urls:
- https://huggingface.co/migtissera/Tess-4.5-27B
- https://huggingface.co/migtissera/Tess-4.5-27B-GGUF
tags:
- llm
- gguf
- cpu
- gpu
- coding
- reasoning
- vision
- multimodal
files:
- filename: Tess-4.5-27B-Q4_K_M.gguf
sha256: 7a3157e2d4ab96a2abcee99a986bea11cfe282b58815ec676c5abe7e6a92d8dc
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/Tess-4.5-27B-Q4_K_M.gguf
- filename: mmproj-Tess-4.5-27B-F16.gguf
sha256: ae93dc35b2bf3a7730fa518db6f1e8d7c624363643f61d85eef8655b671257bb
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/mmproj-Tess-4.5-27B-F16.gguf
overrides:
backend: llama-cpp
context_size: 32768
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: mmproj-Tess-4.5-27B-F16.gguf
options:
- use_jinja:true
parameters:
model: Tess-4.5-27B-Q4_K_M.gguf
temperature: 1.0
top_k: 20
top_p: 0.95
template:
use_tokenizer_template: true
variants:
- model: tess-4.5-27b-q6
- model: tess-4.5-27b-q8
- url: github:mudler/LocalAI/gallery/virtual.yaml@master
name: tess-4.5-27b-q6
description: |
Tess 4.5 is a 27B Qwen3.8 fine-tune for coding, reasoning, and tool use.
This Q6_K GGUF build includes the F16 vision projector and uses
llama.cpp with the embedded chat template. Speculative decoding is disabled.
license: apache-2.0
urls:
- https://huggingface.co/migtissera/Tess-4.5-27B
- https://huggingface.co/migtissera/Tess-4.5-27B-GGUF
tags:
- llm
- gguf
- cpu
- gpu
- coding
- reasoning
- vision
- multimodal
files:
- filename: Tess-4.5-27B-Q6_K.gguf
sha256: 515a2a0339bbb6fa9083cf347755c1a08acb735a43ec1e20f6c361da9491f43b
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/Tess-4.5-27B-Q6_K.gguf
- filename: mmproj-Tess-4.5-27B-F16.gguf
sha256: ae93dc35b2bf3a7730fa518db6f1e8d7c624363643f61d85eef8655b671257bb
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/mmproj-Tess-4.5-27B-F16.gguf
overrides:
backend: llama-cpp
context_size: 32768
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: mmproj-Tess-4.5-27B-F16.gguf
options:
- use_jinja:true
parameters:
model: Tess-4.5-27B-Q6_K.gguf
temperature: 1.0
top_k: 20
top_p: 0.95
template:
use_tokenizer_template: true
- url: github:mudler/LocalAI/gallery/virtual.yaml@master
name: tess-4.5-27b-q8
description: |
Tess 4.5 is a 27B Qwen3.8 fine-tune for coding, reasoning, and tool use.
This Q8_0 GGUF build includes the F16 vision projector and uses
llama.cpp with the embedded chat template. Speculative decoding is disabled.
license: apache-2.0
urls:
- https://huggingface.co/migtissera/Tess-4.5-27B
- https://huggingface.co/migtissera/Tess-4.5-27B-GGUF
tags:
- llm
- gguf
- cpu
- gpu
- coding
- reasoning
- vision
- multimodal
files:
- filename: Tess-4.5-27B-Q8_0.gguf
sha256: 5e97564e427b8672c7e69e119534600b29cb1c1677381cd355be3d140c8aae7a
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/Tess-4.5-27B-Q8_0.gguf
- filename: mmproj-Tess-4.5-27B-F16.gguf
sha256: ae93dc35b2bf3a7730fa518db6f1e8d7c624363643f61d85eef8655b671257bb
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/mmproj-Tess-4.5-27B-F16.gguf
overrides:
backend: llama-cpp
context_size: 32768
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: mmproj-Tess-4.5-27B-F16.gguf
options:
- use_jinja:true
parameters:
model: Tess-4.5-27B-Q8_0.gguf
temperature: 1.0
top_k: 20
top_p: 0.95
template:
use_tokenizer_template: true
- url: github:mudler/LocalAI/gallery/virtual.yaml@master
name: adavar-3b-q4
description: |
gallery/models/index.yaml
+165
View File
@@ -4,6 +4,171 @@
# changes and never edit it by hand. CI fails when it is stale. A build from
# master reads it as the default model source; a tagged build reads the gallery
# until the release cutover.
- kind: model
id: tess-4.5-27b-q4
name: tess-4.5-27b-q4
version: 0.0.0-gallery.c7df49b0fa37
description: |
Tess 4.5 is a 27B Qwen3.8 fine-tune for coding, reasoning, and tool use.
This Q4_K_M GGUF build includes the F16 vision projector and uses
llama.cpp with the embedded chat template. Speculative decoding is disabled.
license: apache-2.0
urls:
- https://huggingface.co/migtissera/Tess-4.5-27B
- https://huggingface.co/migtissera/Tess-4.5-27B-GGUF
tags:
- llm
- gguf
- cpu
- gpu
- coding
- reasoning
- vision
- multimodal
channel: gallery
policy:
requireSignature: false
schemaVersion: 1
model:
url: github:mudler/LocalAI/gallery/virtual.yaml@master
files:
- filename: Tess-4.5-27B-Q4_K_M.gguf
sha256: 7a3157e2d4ab96a2abcee99a986bea11cfe282b58815ec676c5abe7e6a92d8dc
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/Tess-4.5-27B-Q4_K_M.gguf
- filename: mmproj-Tess-4.5-27B-F16.gguf
sha256: ae93dc35b2bf3a7730fa518db6f1e8d7c624363643f61d85eef8655b671257bb
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/mmproj-Tess-4.5-27B-F16.gguf
overrides:
backend: llama-cpp
context_size: 32768
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: mmproj-Tess-4.5-27B-F16.gguf
options:
- use_jinja:true
parameters:
model: Tess-4.5-27B-Q4_K_M.gguf
temperature: 1.0
top_k: 20
top_p: 0.95
template:
use_tokenizer_template: true
variants:
- model: tess-4.5-27b-q6
- model: tess-4.5-27b-q8
- kind: model
id: tess-4.5-27b-q6
name: tess-4.5-27b-q6
version: 0.0.0-gallery.636b28a09c32
description: |
Tess 4.5 is a 27B Qwen3.8 fine-tune for coding, reasoning, and tool use.
This Q6_K GGUF build includes the F16 vision projector and uses
llama.cpp with the embedded chat template. Speculative decoding is disabled.
license: apache-2.0
urls:
- https://huggingface.co/migtissera/Tess-4.5-27B
- https://huggingface.co/migtissera/Tess-4.5-27B-GGUF
tags:
- llm
- gguf
- cpu
- gpu
- coding
- reasoning
- vision
- multimodal
channel: gallery
policy:
requireSignature: false
schemaVersion: 1
model:
url: github:mudler/LocalAI/gallery/virtual.yaml@master
files:
- filename: Tess-4.5-27B-Q6_K.gguf
sha256: 515a2a0339bbb6fa9083cf347755c1a08acb735a43ec1e20f6c361da9491f43b
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/Tess-4.5-27B-Q6_K.gguf
- filename: mmproj-Tess-4.5-27B-F16.gguf
sha256: ae93dc35b2bf3a7730fa518db6f1e8d7c624363643f61d85eef8655b671257bb
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/mmproj-Tess-4.5-27B-F16.gguf
overrides:
backend: llama-cpp
context_size: 32768
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: mmproj-Tess-4.5-27B-F16.gguf
options:
- use_jinja:true
parameters:
model: Tess-4.5-27B-Q6_K.gguf
temperature: 1.0
top_k: 20
top_p: 0.95
template:
use_tokenizer_template: true
- kind: model
id: tess-4.5-27b-q8
name: tess-4.5-27b-q8
version: 0.0.0-gallery.b120ec5f9e63
description: |
Tess 4.5 is a 27B Qwen3.8 fine-tune for coding, reasoning, and tool use.
This Q8_0 GGUF build includes the F16 vision projector and uses
llama.cpp with the embedded chat template. Speculative decoding is disabled.
license: apache-2.0
urls:
- https://huggingface.co/migtissera/Tess-4.5-27B
- https://huggingface.co/migtissera/Tess-4.5-27B-GGUF
tags:
- llm
- gguf
- cpu
- gpu
- coding
- reasoning
- vision
- multimodal
channel: gallery
policy:
requireSignature: false
schemaVersion: 1
model:
url: github:mudler/LocalAI/gallery/virtual.yaml@master
files:
- filename: Tess-4.5-27B-Q8_0.gguf
sha256: 5e97564e427b8672c7e69e119534600b29cb1c1677381cd355be3d140c8aae7a
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/Tess-4.5-27B-Q8_0.gguf
- filename: mmproj-Tess-4.5-27B-F16.gguf
sha256: ae93dc35b2bf3a7730fa518db6f1e8d7c624363643f61d85eef8655b671257bb
uri: https://huggingface.co/migtissera/Tess-4.5-27B-GGUF/resolve/af17a3f0de2b8557849c1b6fcc303d3160052450/mmproj-Tess-4.5-27B-F16.gguf
overrides:
backend: llama-cpp
context_size: 32768
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: mmproj-Tess-4.5-27B-F16.gguf
options:
- use_jinja:true
parameters:
model: Tess-4.5-27B-Q8_0.gguf
temperature: 1.0
top_k: 20
top_p: 0.95
template:
use_tokenizer_template: true
- kind: model
id: adavar-3b-q4
name: adavar-3b-q4