mirror of
https://github.com/mudler/LocalAI.git
synced 2026-10-28 00:59:28 -04:00
feat(gallery): add Diplodocus preview variants
Add Q4_K_M, Q6_K, and Q8_0 builds for llama.cpp with pinned assets. Group the builds and document installation, reasoning, and tool setup. Assisted-by: Codex:gpt-6
This commit is contained in:
localai-org-maint-bot committed
1 parent
0a44303c21
commit
f9e7f0bfbe
3 files changed
+289
No files matched your search
@@ -61,6 +61,25 @@ Select a specific build with `--variant minicpm-v-4_7-1b-q4`,
|
||||
`--variant minicpm-v-4_7-1b-q5`, `--variant minicpm-v-4_7-1b-q6`,
|
||||
or `--variant minicpm-v-4_7-1b-q8`.
|
||||
|
||||
### Diplodocus preview
|
||||
|
||||
[Diplodocus preview 4](https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF)
|
||||
is a K2 Horizon 7B fine-tune for documents, spreadsheets, and tool-assisted tasks.
|
||||
The gallery provides Q4_K_M, Q6_K, and Q8_0 builds for llama.cpp:
|
||||
|
||||
```bash
|
||||
local-ai models install --variant diplodocus-preview-q6 diplodocus-preview-q4
|
||||
```
|
||||
|
||||
Use `diplodocus-preview-q4` or `diplodocus-preview-q8` as the variant to select
|
||||
another size. These builds use the embedded chat template, a 32,768-token
|
||||
context, temperature 1.0, and top-p 0.95.
|
||||
|
||||
Reasoning defaults to `high`. Set `reasoning_effort` to `low` in a chat request
|
||||
to disable reasoning. The publisher recommends `high` or `low` and advises against `medium`.
|
||||
Your application must define and execute tools such as `calculate`; installing
|
||||
the model does not install the publisher's document or spreadsheet tools.
|
||||
|
||||
### Repairing an existing model configuration
|
||||
|
||||
Gallery template updates do not rewrite YAML files for already installed models.
|
||||
|
||||
@@ -3,6 +3,129 @@
|
||||
# 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: diplodocus-preview-q4
|
||||
description: |
|
||||
Diplodocus preview 4 is OsaurusAI's K2 Horizon 7B fine-tune for documents,
|
||||
spreadsheets, and tool-assisted tasks. This Q4_K_M GGUF build uses the
|
||||
embedded chat template and enables reasoning by default. Applications
|
||||
must provide the tools; the model download does not include workspace tools.
|
||||
license: apache-2.0
|
||||
urls:
|
||||
- https://huggingface.co/IFM/K2-Horizon-7B
|
||||
- https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF
|
||||
tags:
|
||||
- llm
|
||||
- gguf
|
||||
- reasoning
|
||||
- function-calling
|
||||
- cpu
|
||||
- gpu
|
||||
files:
|
||||
- filename: Diplodocus-preview-Q4_K_M.gguf
|
||||
sha256: 32c4d81cacb4af48c61e852e8bca2d974548900aca337088c14c2869e15311e4
|
||||
uri: https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF/resolve/7072381575adfe4e528349799f30d27488a4ca89/Diplodocus-preview-Q4_K_M.gguf
|
||||
overrides:
|
||||
backend: llama-cpp
|
||||
context_size: 32768
|
||||
function:
|
||||
automatic_tool_parsing_fallback: true
|
||||
grammar:
|
||||
disable: true
|
||||
known_usecases:
|
||||
- chat
|
||||
options:
|
||||
- use_jinja:true
|
||||
parameters:
|
||||
model: Diplodocus-preview-Q4_K_M.gguf
|
||||
temperature: 1.0
|
||||
top_p: 0.95
|
||||
reasoning_effort: high
|
||||
template:
|
||||
use_tokenizer_template: true
|
||||
variants:
|
||||
- model: diplodocus-preview-q6
|
||||
- model: diplodocus-preview-q8
|
||||
- url: github:mudler/LocalAI/gallery/virtual.yaml@master
|
||||
name: diplodocus-preview-q6
|
||||
description: |
|
||||
Diplodocus preview 4 is OsaurusAI's K2 Horizon 7B fine-tune for documents,
|
||||
spreadsheets, and tool-assisted tasks. This Q6_K GGUF build uses the
|
||||
embedded chat template and enables reasoning by default. Applications
|
||||
must provide the tools; the model download does not include workspace tools.
|
||||
license: apache-2.0
|
||||
urls:
|
||||
- https://huggingface.co/IFM/K2-Horizon-7B
|
||||
- https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF
|
||||
tags:
|
||||
- llm
|
||||
- gguf
|
||||
- reasoning
|
||||
- function-calling
|
||||
- cpu
|
||||
- gpu
|
||||
files:
|
||||
- filename: Diplodocus-preview-Q6_K.gguf
|
||||
sha256: c81c80cc2f3d5f9fec0eb06d9a8732404ac37516fde2d8692eb49f6e1d39cfd9
|
||||
uri: https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF/resolve/7072381575adfe4e528349799f30d27488a4ca89/Diplodocus-preview-Q6_K.gguf
|
||||
overrides:
|
||||
backend: llama-cpp
|
||||
context_size: 32768
|
||||
function:
|
||||
automatic_tool_parsing_fallback: true
|
||||
grammar:
|
||||
disable: true
|
||||
known_usecases:
|
||||
- chat
|
||||
options:
|
||||
- use_jinja:true
|
||||
parameters:
|
||||
model: Diplodocus-preview-Q6_K.gguf
|
||||
temperature: 1.0
|
||||
top_p: 0.95
|
||||
reasoning_effort: high
|
||||
template:
|
||||
use_tokenizer_template: true
|
||||
- url: github:mudler/LocalAI/gallery/virtual.yaml@master
|
||||
name: diplodocus-preview-q8
|
||||
description: |
|
||||
Diplodocus preview 4 is OsaurusAI's K2 Horizon 7B fine-tune for documents,
|
||||
spreadsheets, and tool-assisted tasks. This Q8_0 GGUF build uses the
|
||||
embedded chat template and enables reasoning by default. Applications
|
||||
must provide the tools; the model download does not include workspace tools.
|
||||
license: apache-2.0
|
||||
urls:
|
||||
- https://huggingface.co/IFM/K2-Horizon-7B
|
||||
- https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF
|
||||
tags:
|
||||
- llm
|
||||
- gguf
|
||||
- reasoning
|
||||
- function-calling
|
||||
- cpu
|
||||
- gpu
|
||||
files:
|
||||
- filename: Diplodocus-preview-Q8_0.gguf
|
||||
sha256: 8d39a2d47b56b308d60d79efa194a2f0796571b439b14499cd4847ccacdf8a16
|
||||
uri: https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF/resolve/7072381575adfe4e528349799f30d27488a4ca89/Diplodocus-preview-Q8_0.gguf
|
||||
overrides:
|
||||
backend: llama-cpp
|
||||
context_size: 32768
|
||||
function:
|
||||
automatic_tool_parsing_fallback: true
|
||||
grammar:
|
||||
disable: true
|
||||
known_usecases:
|
||||
- chat
|
||||
options:
|
||||
- use_jinja:true
|
||||
parameters:
|
||||
model: Diplodocus-preview-Q8_0.gguf
|
||||
temperature: 1.0
|
||||
top_p: 0.95
|
||||
reasoning_effort: high
|
||||
template:
|
||||
use_tokenizer_template: true
|
||||
- url: github:mudler/LocalAI/gallery/virtual.yaml@master
|
||||
name: adavar-3b-q4
|
||||
description: |
|
||||
|
||||
@@ -4,6 +4,153 @@
|
||||
# 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: diplodocus-preview-q4
|
||||
name: diplodocus-preview-q4
|
||||
version: 0.0.0-gallery.fe37350f336e
|
||||
description: |
|
||||
Diplodocus preview 4 is OsaurusAI's K2 Horizon 7B fine-tune for documents,
|
||||
spreadsheets, and tool-assisted tasks. This Q4_K_M GGUF build uses the
|
||||
embedded chat template and enables reasoning by default. Applications
|
||||
must provide the tools; the model download does not include workspace tools.
|
||||
license: apache-2.0
|
||||
urls:
|
||||
- https://huggingface.co/IFM/K2-Horizon-7B
|
||||
- https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF
|
||||
tags:
|
||||
- llm
|
||||
- gguf
|
||||
- reasoning
|
||||
- function-calling
|
||||
- cpu
|
||||
- gpu
|
||||
channel: gallery
|
||||
policy:
|
||||
requireSignature: false
|
||||
schemaVersion: 1
|
||||
model:
|
||||
url: github:mudler/LocalAI/gallery/virtual.yaml@master
|
||||
files:
|
||||
- filename: Diplodocus-preview-Q4_K_M.gguf
|
||||
sha256: 32c4d81cacb4af48c61e852e8bca2d974548900aca337088c14c2869e15311e4
|
||||
uri: https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF/resolve/7072381575adfe4e528349799f30d27488a4ca89/Diplodocus-preview-Q4_K_M.gguf
|
||||
overrides:
|
||||
backend: llama-cpp
|
||||
context_size: 32768
|
||||
function:
|
||||
automatic_tool_parsing_fallback: true
|
||||
grammar:
|
||||
disable: true
|
||||
known_usecases:
|
||||
- chat
|
||||
options:
|
||||
- use_jinja:true
|
||||
parameters:
|
||||
model: Diplodocus-preview-Q4_K_M.gguf
|
||||
temperature: 1.0
|
||||
top_p: 0.95
|
||||
reasoning_effort: high
|
||||
template:
|
||||
use_tokenizer_template: true
|
||||
variants:
|
||||
- model: diplodocus-preview-q6
|
||||
- model: diplodocus-preview-q8
|
||||
- kind: model
|
||||
id: diplodocus-preview-q6
|
||||
name: diplodocus-preview-q6
|
||||
version: 0.0.0-gallery.64af0b4c83da
|
||||
description: |
|
||||
Diplodocus preview 4 is OsaurusAI's K2 Horizon 7B fine-tune for documents,
|
||||
spreadsheets, and tool-assisted tasks. This Q6_K GGUF build uses the
|
||||
embedded chat template and enables reasoning by default. Applications
|
||||
must provide the tools; the model download does not include workspace tools.
|
||||
license: apache-2.0
|
||||
urls:
|
||||
- https://huggingface.co/IFM/K2-Horizon-7B
|
||||
- https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF
|
||||
tags:
|
||||
- llm
|
||||
- gguf
|
||||
- reasoning
|
||||
- function-calling
|
||||
- cpu
|
||||
- gpu
|
||||
channel: gallery
|
||||
policy:
|
||||
requireSignature: false
|
||||
schemaVersion: 1
|
||||
model:
|
||||
url: github:mudler/LocalAI/gallery/virtual.yaml@master
|
||||
files:
|
||||
- filename: Diplodocus-preview-Q6_K.gguf
|
||||
sha256: c81c80cc2f3d5f9fec0eb06d9a8732404ac37516fde2d8692eb49f6e1d39cfd9
|
||||
uri: https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF/resolve/7072381575adfe4e528349799f30d27488a4ca89/Diplodocus-preview-Q6_K.gguf
|
||||
overrides:
|
||||
backend: llama-cpp
|
||||
context_size: 32768
|
||||
function:
|
||||
automatic_tool_parsing_fallback: true
|
||||
grammar:
|
||||
disable: true
|
||||
known_usecases:
|
||||
- chat
|
||||
options:
|
||||
- use_jinja:true
|
||||
parameters:
|
||||
model: Diplodocus-preview-Q6_K.gguf
|
||||
temperature: 1.0
|
||||
top_p: 0.95
|
||||
reasoning_effort: high
|
||||
template:
|
||||
use_tokenizer_template: true
|
||||
- kind: model
|
||||
id: diplodocus-preview-q8
|
||||
name: diplodocus-preview-q8
|
||||
version: 0.0.0-gallery.d0f535558239
|
||||
description: |
|
||||
Diplodocus preview 4 is OsaurusAI's K2 Horizon 7B fine-tune for documents,
|
||||
spreadsheets, and tool-assisted tasks. This Q8_0 GGUF build uses the
|
||||
embedded chat template and enables reasoning by default. Applications
|
||||
must provide the tools; the model download does not include workspace tools.
|
||||
license: apache-2.0
|
||||
urls:
|
||||
- https://huggingface.co/IFM/K2-Horizon-7B
|
||||
- https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF
|
||||
tags:
|
||||
- llm
|
||||
- gguf
|
||||
- reasoning
|
||||
- function-calling
|
||||
- cpu
|
||||
- gpu
|
||||
channel: gallery
|
||||
policy:
|
||||
requireSignature: false
|
||||
schemaVersion: 1
|
||||
model:
|
||||
url: github:mudler/LocalAI/gallery/virtual.yaml@master
|
||||
files:
|
||||
- filename: Diplodocus-preview-Q8_0.gguf
|
||||
sha256: 8d39a2d47b56b308d60d79efa194a2f0796571b439b14499cd4847ccacdf8a16
|
||||
uri: https://huggingface.co/OsaurusAI/Diplodocus-preview-GGUF/resolve/7072381575adfe4e528349799f30d27488a4ca89/Diplodocus-preview-Q8_0.gguf
|
||||
overrides:
|
||||
backend: llama-cpp
|
||||
context_size: 32768
|
||||
function:
|
||||
automatic_tool_parsing_fallback: true
|
||||
grammar:
|
||||
disable: true
|
||||
known_usecases:
|
||||
- chat
|
||||
options:
|
||||
- use_jinja:true
|
||||
parameters:
|
||||
model: Diplodocus-preview-Q8_0.gguf
|
||||
temperature: 1.0
|
||||
top_p: 0.95
|
||||
reasoning_effort: high
|
||||
template:
|
||||
use_tokenizer_template: true
|
||||
- kind: model
|
||||
id: adavar-3b-q4
|
||||
name: adavar-3b-q4
|
||||
|
||||
Reference in new issue
Block a user